|
On June 13 2012 02:03 cLutZ wrote:Show nested quote +On June 12 2012 22:45 BlueSpace wrote: 1) Cause and effect:
The OP has presented data which shows that a 10% gold lead by 12 minutes leads in 90% of the cases to win for the leading team. People here interpret this in a way which indicates that the rest of the game is "meaningless". You could also interpret this in way that the stronger team will manage to pull ahead early and this shows.
This is...false. Randomness decreases as gametime increases. It also decreases as the number of decisions a team, as a whole, makes. The problem? Early game is the part of the game where teams make the fewest real decisions. In reality, a few (think 3-4) are turning the outcome of the game. Particularly in pro games. Last hitting is pretty standard, so that isn't going to be a huge advantage (unless you got a kill, which is what probably happened to create this lead 1 kill, or 2), team comp could be giving you an advantage (your jungler could be faster), but usually its a bunch of RNG and highly guess-y things determining who has the lead at this point. Good examples of what I'm saying: A successful or failed invade; a successful gank, perhaps 2; a countergank, or bait bottom. Most of these are based on where the jungler is, and are very dependent on the fog of war, and getting good ward placement at this point in the game is nearly impossible. So, not only are you determining the outcome of the game during a small portion of the game, you are also determining the outcome during the period of the game that is most fundamentally random. Thats why you see so many safe picks in pro games, particularly top and mid.
1) The game start is not random at all. Everything including buffs are fixed at the game start. You have zero randomness at the beginning of the game, therefor randomness has to increase as the game progresses and cannot constantly decrease.
2) Last hitting at pro level is only standard in so far, that when left alone everybody is pretty perfect at it. In reality you have the enemy jungler and the enemy laner to worry about. The ability to last hit is not entirely decided inside the game but heavily influenced by the picks and bans prior to game start. Getting outpicked will lead to a bad lane matchup which in turn will lead to less CS as the enemy laner can harass more effectively. There is nothing random about that.
3) Jungling is not random. The jungler does not randomly run around the map and ganks lanes. He can't because he will fall behind in farm. Positioning of the jungler is usually quite well understood and there are specific timings for ganks. Pro players know when a lvl 2 gank might happen for example. The counter is to not overextend. Not Random...
4) Ward placement in the early game is not hard. Everybody is in their lane. You know from where the enemy jungler will come and all towers are up. Warding late game is much harder since you might be down several towers and at any given time several people might be missing from their lanes and lurking in the bushes, waiting to attack you. Again no randomness...
I don't want to say that there is no luck involved in LoL, but there is no inherent randomness to the early game it is as skill and comp based as any other part of the game.
|
Please get your wording right. There is no such thing as "randomness" in a game like League, SC, DotA or Poker. The closest you can get to "random" is stuff like crit chance and that isn't really random either.
The amount of variables which lead to results increases over a certain amount of time and will then decrease again towards the end. In practice, certain actions from the enemy are more likely than others (it's almost impossible to truly randomize your actions in a game like Poker, it is literally impossible to do so in a game of League) - whoever reads the limited amount of information he has better will make more sound decisions.
Ingame you don't fight the actual actions of the enemy, you fight against a possible range of actions (simply because you can't know what anyone else will do next but you can try to guess it based on the information you have).
|
On June 14 2012 14:23 r.Evo wrote:Show nested quote +On June 14 2012 14:01 sylverfyre wrote: I'm curious about what happens when you throw out all the 2-0 series as "one team was TOO much better than the other team, and is obviously going to win before the game is even started"
The core of the issue is not that we're interested in how likely a comeback is when one team is significantly worse than another. We need to look at what happens when the teams are at least reasonably close in skill. TSM is always going to stomp a team that is far worse than they are, even within the first 12 minutes. It might happen that they botch early game and make a comeback, but that requires them to somehow lose the early game against a significantly worse team - it's not going to happen often.
That phenomenon isn't exclusive to LoL either - It's not very interesting when you have the #1 team in any sport stomping some of the lower teams in teh league, and leading by a huge margin halfway through the game (I love you Cubbies, but I'm looking at you here.) TSM vs CLG comebacks are what make watching LoL interesting, and those are not limited to 10% of games between top teams. Need to measure more against even teams. While we can't necessarily rate skill numerically, we CAN say where it's "roughly even", such as comparing numbers in matchups where the teams have relatively even win/loss rate against each other. Now you're trying to figure out how to skew the numbers. =P Take the top 5 for a specific tournament, only count games they played vs each other, do that for a time period of one year. That'd be as close as you could get to "even skill" while maintaining a somewhat healthy sample size. No, I'm trying to look at a subset of the data that throws out the clutter where you have one team absolutely stomping 50% gold lead in under 20 minutes because they outclass the other so badly it's not even funny. It's the equivalent of having a pro team vs an amateur team in any other sport- if they DIDN'T absolutely crush the opposition, you'd be impressed by the amateur team's ability to up up a fight - but if you put it into the dataset the same way, you're giving it the same weight as a TSM vs CLG game, when one is obviously meaningful to the "snowballing 2 strong" conclusion and the other is not.
On June 14 2012 18:15 r.Evo wrote: Please get your wording right. There is no such thing as "randomness" in a game like League, SC, DotA or Poker. The closest you can get to "random" is stuff like crit chance and that isn't really random either.
The amount of variables which lead to results increases over a certain amount of time and will then decrease again towards the end. In practice, certain actions from the enemy are more likely than others (it's almost impossible to truly randomize your actions in a game like Poker, it is literally impossible to do so in a game of League) - whoever reads the limited amount of information he has better will make more sound decisions.
Ingame you don't fight the actual actions of the enemy, you fight against a possible range of actions (simply because you can't know what anyone else will do next but you can try to guess it based on the information you have). Get your own wording right before you call people out. Yes, there's randomness in poker. There is in DotA too. Critting and phage-proccing in LoL are basically the only random events, but it's still random. And whenever you're making decisions based off of incomplete information, (All games described) you can basically treat
The difference is, a smart player can narrow down the possible outcomes of a random/unknown event (in poker, think of pinning your opponent on a particular range of hands) and often prepare for or at least EXPECT the remaining possible [random or simply unknown] possibilities.
|
The "randomness" you're describing doesn't exist if you look at the decisionmaking process. And that is the process which matters. The actual outcome has a very slight random element but that's not what we're talking about.
The "random factor" you're talking about is plain and simple ones inability to narrow down the possibilites due to limited information. Not sure why you call me out on that when I even wrote myself that crit chance is as close as you can get to something random. As close as you can get, because it is not random.
Both DotA and League of Legends use a pseudo-random distribution for crit-chance, here's an explanation of the old DotA one, not sure what changed lately: http://www.dotastrategy.com/forum/ftopic18287.html ... LoL uses a similar, modified, system but it's also not really random.
The difference is, a smart player can narrow down the possible outcomes of a random/unknown event (in poker, think of pinning your opponent on a particular range of hands) and often prepare for or at least EXPECT the remaining possible [random or simply unknown] possibilities. What you're describing is not about random events. The good player understands the possibilites that are out there, narrows the opponents behaviour down to ranges and then makes a profitable decision based on those ranges. That eliminates the random factor in the long run completely. Yes, completely.
Sure, it can happen that a certain player makes a huge misclick which ends up with him in a clever spot no one involved thought about before which then leads to an unaccounted kill. But that isn't about randomness in the design of the game.
An actual example for randomness via design would be Chaos Knight from DotA. =P
|
On June 14 2012 17:44 BlueSpace wrote:Show nested quote +On June 13 2012 02:03 cLutZ wrote:On June 12 2012 22:45 BlueSpace wrote: 1) Cause and effect:
The OP has presented data which shows that a 10% gold lead by 12 minutes leads in 90% of the cases to win for the leading team. People here interpret this in a way which indicates that the rest of the game is "meaningless". You could also interpret this in way that the stronger team will manage to pull ahead early and this shows.
This is...false. Randomness decreases as gametime increases. It also decreases as the number of decisions a team, as a whole, makes. The problem? Early game is the part of the game where teams make the fewest real decisions. In reality, a few (think 3-4) are turning the outcome of the game. Particularly in pro games. Last hitting is pretty standard, so that isn't going to be a huge advantage (unless you got a kill, which is what probably happened to create this lead 1 kill, or 2), team comp could be giving you an advantage (your jungler could be faster), but usually its a bunch of RNG and highly guess-y things determining who has the lead at this point. Good examples of what I'm saying: A successful or failed invade; a successful gank, perhaps 2; a countergank, or bait bottom. Most of these are based on where the jungler is, and are very dependent on the fog of war, and getting good ward placement at this point in the game is nearly impossible. So, not only are you determining the outcome of the game during a small portion of the game, you are also determining the outcome during the period of the game that is most fundamentally random. Thats why you see so many safe picks in pro games, particularly top and mid. 1) The game start is not random at all. Everything including buffs are fixed at the game start. You have zero randomness at the beginning of the game, therefor randomness has to increase as the game progresses and cannot constantly decrease. 2) Last hitting at pro level is only standard in so far, that when left alone everybody is pretty perfect at it. In reality you have the enemy jungler and the enemy laner to worry about. The ability to last hit is not entirely decided inside the game but heavily influenced by the picks and bans prior to game start. Getting outpicked will lead to a bad lane matchup which in turn will lead to less CS as the enemy laner can harass more effectively. There is nothing random about that. 3) Jungling is not random. The jungler does not randomly run around the map and ganks lanes. He can't because he will fall behind in farm. Positioning of the jungler is usually quite well understood and there are specific timings for ganks. Pro players know when a lvl 2 gank might happen for example. The counter is to not overextend. Not Random... 4) Ward placement in the early game is not hard. Everybody is in their lane. You know from where the enemy jungler will come and all towers are up. Warding late game is much harder since you might be down several towers and at any given time several people might be missing from their lanes and lurking in the bushes, waiting to attack you. Again no randomness... I don't want to say that there is no luck involved in LoL, but there is no inherent randomness to the early game it is as skill and comp based as any other part of the game.
Random was, of course, the wrong word to use. What I should have said is "uncertainty". 1. True. 2. Not relevant to this, except to the point that this reinforces my point that the snowballiness of the game significantly reduces the number of viable champions and team compositions. 3. Again, not random, uncertainty. You can't control where the jungler is, and, particularly top lane, can't have enough wards to protect itself. 4. Warding late game is easier because you have more gold and you are grouped as a team so warding is less necessary.
You are basically saying that caution=skill. And in the laning phase that is almost entirely correct. Janna throwing a tornado into a bush, even if no one is there, is better off wasting both the mana and the CD than not and saving it for an aggressive play. I'm saying that this means a bunch of other things that are bad. DO you remember when M5 was super dominant with their jungle Alistar stuff? Guess what, it was based on this basic principle: get an early kill or 2 = win game.
On June 14 2012 23:29 r.Evo wrote:The "randomness" you're describing doesn't exist if you look at the decisionmaking process. And that is the process which matters. The actual outcome has a very slight random element but that's not what we're talking about. The "random factor" you're talking about is plain and simple ones inability to narrow down the possibilites due to limited information. Not sure why you call me out on that when I even wrote myself that crit chance is as close as you can get to something random. As close as you can get, because it is not random. Both DotA and League of Legends use a pseudo-random distribution for crit-chance, here's an explanation of the old DotA one, not sure what changed lately: http://www.dotastrategy.com/forum/ftopic18287.html ... LoL uses a similar, modified, system but it's also not really random. Show nested quote +The difference is, a smart player can narrow down the possible outcomes of a random/unknown event (in poker, think of pinning your opponent on a particular range of hands) and often prepare for or at least EXPECT the remaining possible [random or simply unknown] possibilities. What you're describing is not about random events. The good player understands the possibilites that are out there, narrows the opponents behaviour down to ranges and then makes a profitable decision based on those ranges. That eliminates the random factor in the long run completely. Yes, completely.Sure, it can happen that a certain player makes a huge misclick which ends up with him in a clever spot no one involved thought about before which then leads to an unaccounted kill. But that isn't about randomness in the design of the game.An actual example for randomness via design would be Chaos Knight from DotA. =P
I underlined the important portion of your text. IN THE LONG RUN. Let us assume that the entire game = the long run. 12 minutes is certainly NOT the long run. What the stats say, however, is if you are making decisions that are risky, but very profitable in theory, you will likely lose.
------
The overall point is, the game needs to be structured in such a way that the sum of the decisions throughout the game is reflected in the final result. Right now, it appears to have a model where that is less likely, and rather the decisions in the beginning of the game have more weight than those later.
|
This is a fundamental game design problem. If you want to avoid it, you need to have kills and towers to not give any gameplay advantage, and points instead like in sports. the whole concept of farming makes this impossible, as competitive farming is all about getting a comparitive advantage. lol would need to be made into a purely PvP game based on score then
|
On June 15 2012 05:16 cLutZ wrote:Show nested quote +On June 14 2012 17:44 BlueSpace wrote:On June 13 2012 02:03 cLutZ wrote:On June 12 2012 22:45 BlueSpace wrote: 1) Cause and effect:
The OP has presented data which shows that a 10% gold lead by 12 minutes leads in 90% of the cases to win for the leading team. People here interpret this in a way which indicates that the rest of the game is "meaningless". You could also interpret this in way that the stronger team will manage to pull ahead early and this shows.
This is...false. Randomness decreases as gametime increases. It also decreases as the number of decisions a team, as a whole, makes. The problem? Early game is the part of the game where teams make the fewest real decisions. In reality, a few (think 3-4) are turning the outcome of the game. Particularly in pro games. Last hitting is pretty standard, so that isn't going to be a huge advantage (unless you got a kill, which is what probably happened to create this lead 1 kill, or 2), team comp could be giving you an advantage (your jungler could be faster), but usually its a bunch of RNG and highly guess-y things determining who has the lead at this point. Good examples of what I'm saying: A successful or failed invade; a successful gank, perhaps 2; a countergank, or bait bottom. Most of these are based on where the jungler is, and are very dependent on the fog of war, and getting good ward placement at this point in the game is nearly impossible. So, not only are you determining the outcome of the game during a small portion of the game, you are also determining the outcome during the period of the game that is most fundamentally random. Thats why you see so many safe picks in pro games, particularly top and mid. 1) The game start is not random at all. Everything including buffs are fixed at the game start. You have zero randomness at the beginning of the game, therefor randomness has to increase as the game progresses and cannot constantly decrease. 2) Last hitting at pro level is only standard in so far, that when left alone everybody is pretty perfect at it. In reality you have the enemy jungler and the enemy laner to worry about. The ability to last hit is not entirely decided inside the game but heavily influenced by the picks and bans prior to game start. Getting outpicked will lead to a bad lane matchup which in turn will lead to less CS as the enemy laner can harass more effectively. There is nothing random about that. 3) Jungling is not random. The jungler does not randomly run around the map and ganks lanes. He can't because he will fall behind in farm. Positioning of the jungler is usually quite well understood and there are specific timings for ganks. Pro players know when a lvl 2 gank might happen for example. The counter is to not overextend. Not Random... 4) Ward placement in the early game is not hard. Everybody is in their lane. You know from where the enemy jungler will come and all towers are up. Warding late game is much harder since you might be down several towers and at any given time several people might be missing from their lanes and lurking in the bushes, waiting to attack you. Again no randomness... I don't want to say that there is no luck involved in LoL, but there is no inherent randomness to the early game it is as skill and comp based as any other part of the game. Random was, of course, the wrong word to use. What I should have said is "uncertainty". 1. True. 2. Not relevant to this, except to the point that this reinforces my point that the snowballiness of the game significantly reduces the number of viable champions and team compositions. 3. Again, not random, uncertainty. You can't control where the jungler is, and, particularly top lane, can't have enough wards to protect itself. 4. Warding late game is easier because you have more gold and you are grouped as a team so warding is less necessary. You are basically saying that caution=skill. And in the laning phase that is almost entirely correct. Janna throwing a tornado into a bush, even if no one is there, is better off wasting both the mana and the CD than not and saving it for an aggressive play. I'm saying that this means a bunch of other things that are bad. DO you remember when M5 was super dominant with their jungle Alistar stuff? Guess what, it was based on this basic principle: get an early kill or 2 = win game. Show nested quote +On June 14 2012 23:29 r.Evo wrote:The "randomness" you're describing doesn't exist if you look at the decisionmaking process. And that is the process which matters. The actual outcome has a very slight random element but that's not what we're talking about. The "random factor" you're talking about is plain and simple ones inability to narrow down the possibilites due to limited information. Not sure why you call me out on that when I even wrote myself that crit chance is as close as you can get to something random. As close as you can get, because it is not random. Both DotA and League of Legends use a pseudo-random distribution for crit-chance, here's an explanation of the old DotA one, not sure what changed lately: http://www.dotastrategy.com/forum/ftopic18287.html ... LoL uses a similar, modified, system but it's also not really random. The difference is, a smart player can narrow down the possible outcomes of a random/unknown event (in poker, think of pinning your opponent on a particular range of hands) and often prepare for or at least EXPECT the remaining possible [random or simply unknown] possibilities. What you're describing is not about random events. The good player understands the possibilites that are out there, narrows the opponents behaviour down to ranges and then makes a profitable decision based on those ranges. That eliminates the random factor in the long run completely. Yes, completely.Sure, it can happen that a certain player makes a huge misclick which ends up with him in a clever spot no one involved thought about before which then leads to an unaccounted kill. But that isn't about randomness in the design of the game.An actual example for randomness via design would be Chaos Knight from DotA. =P I underlined the important portion of your text. IN THE LONG RUN. Let us assume that the entire game = the long run. 12 minutes is certainly NOT the long run. What the stats say, however, is if you are making decisions that are risky, but very profitable in theory, you will likely lose. ------ The overall point is, the game needs to be structured in such a way that the sum of the decisions throughout the game is reflected in the final result. Right now, it appears to have a model where that is less likely, and rather the decisions in the beginning of the game have more weight than those later. The problem is, this WILL NOT HAPPEN. Why? Because THAT DOESNT SELL. Look at the games that have become the most popular. They are all farming games. Farming/grinding to get a comparitive advantage over an enemy. CS, WoW, CoD(rank farming), LoL, Farmville. ALL the most popular games these days are about farming. And farming is in all its implementations snowbally: the winning farmer becomes better than the losing farmer. Why is this always made snowbally? because non-snowbally farming IS NOT FUN. There is little motivation to win(get a kill, etc) if you know your opponent will gain just as much from it as you do.
In short, If a game has farming in it(as popular modern games do), it will be snowbally. Thats all there is to it. You would need to have like an ARAM with gold only from timer to make it linear
|
On June 15 2012 07:38 brolaf wrote:Show nested quote +On June 15 2012 05:16 cLutZ wrote:On June 14 2012 17:44 BlueSpace wrote:On June 13 2012 02:03 cLutZ wrote:On June 12 2012 22:45 BlueSpace wrote: 1) Cause and effect:
The OP has presented data which shows that a 10% gold lead by 12 minutes leads in 90% of the cases to win for the leading team. People here interpret this in a way which indicates that the rest of the game is "meaningless". You could also interpret this in way that the stronger team will manage to pull ahead early and this shows.
This is...false. Randomness decreases as gametime increases. It also decreases as the number of decisions a team, as a whole, makes. The problem? Early game is the part of the game where teams make the fewest real decisions. In reality, a few (think 3-4) are turning the outcome of the game. Particularly in pro games. Last hitting is pretty standard, so that isn't going to be a huge advantage (unless you got a kill, which is what probably happened to create this lead 1 kill, or 2), team comp could be giving you an advantage (your jungler could be faster), but usually its a bunch of RNG and highly guess-y things determining who has the lead at this point. Good examples of what I'm saying: A successful or failed invade; a successful gank, perhaps 2; a countergank, or bait bottom. Most of these are based on where the jungler is, and are very dependent on the fog of war, and getting good ward placement at this point in the game is nearly impossible. So, not only are you determining the outcome of the game during a small portion of the game, you are also determining the outcome during the period of the game that is most fundamentally random. Thats why you see so many safe picks in pro games, particularly top and mid. 1) The game start is not random at all. Everything including buffs are fixed at the game start. You have zero randomness at the beginning of the game, therefor randomness has to increase as the game progresses and cannot constantly decrease. 2) Last hitting at pro level is only standard in so far, that when left alone everybody is pretty perfect at it. In reality you have the enemy jungler and the enemy laner to worry about. The ability to last hit is not entirely decided inside the game but heavily influenced by the picks and bans prior to game start. Getting outpicked will lead to a bad lane matchup which in turn will lead to less CS as the enemy laner can harass more effectively. There is nothing random about that. 3) Jungling is not random. The jungler does not randomly run around the map and ganks lanes. He can't because he will fall behind in farm. Positioning of the jungler is usually quite well understood and there are specific timings for ganks. Pro players know when a lvl 2 gank might happen for example. The counter is to not overextend. Not Random... 4) Ward placement in the early game is not hard. Everybody is in their lane. You know from where the enemy jungler will come and all towers are up. Warding late game is much harder since you might be down several towers and at any given time several people might be missing from their lanes and lurking in the bushes, waiting to attack you. Again no randomness... I don't want to say that there is no luck involved in LoL, but there is no inherent randomness to the early game it is as skill and comp based as any other part of the game. Random was, of course, the wrong word to use. What I should have said is "uncertainty". 1. True. 2. Not relevant to this, except to the point that this reinforces my point that the snowballiness of the game significantly reduces the number of viable champions and team compositions. 3. Again, not random, uncertainty. You can't control where the jungler is, and, particularly top lane, can't have enough wards to protect itself. 4. Warding late game is easier because you have more gold and you are grouped as a team so warding is less necessary. You are basically saying that caution=skill. And in the laning phase that is almost entirely correct. Janna throwing a tornado into a bush, even if no one is there, is better off wasting both the mana and the CD than not and saving it for an aggressive play. I'm saying that this means a bunch of other things that are bad. DO you remember when M5 was super dominant with their jungle Alistar stuff? Guess what, it was based on this basic principle: get an early kill or 2 = win game. On June 14 2012 23:29 r.Evo wrote:The "randomness" you're describing doesn't exist if you look at the decisionmaking process. And that is the process which matters. The actual outcome has a very slight random element but that's not what we're talking about. The "random factor" you're talking about is plain and simple ones inability to narrow down the possibilites due to limited information. Not sure why you call me out on that when I even wrote myself that crit chance is as close as you can get to something random. As close as you can get, because it is not random. Both DotA and League of Legends use a pseudo-random distribution for crit-chance, here's an explanation of the old DotA one, not sure what changed lately: http://www.dotastrategy.com/forum/ftopic18287.html ... LoL uses a similar, modified, system but it's also not really random. The difference is, a smart player can narrow down the possible outcomes of a random/unknown event (in poker, think of pinning your opponent on a particular range of hands) and often prepare for or at least EXPECT the remaining possible [random or simply unknown] possibilities. What you're describing is not about random events. The good player understands the possibilites that are out there, narrows the opponents behaviour down to ranges and then makes a profitable decision based on those ranges. That eliminates the random factor in the long run completely. Yes, completely.Sure, it can happen that a certain player makes a huge misclick which ends up with him in a clever spot no one involved thought about before which then leads to an unaccounted kill. But that isn't about randomness in the design of the game.An actual example for randomness via design would be Chaos Knight from DotA. =P I underlined the important portion of your text. IN THE LONG RUN. Let us assume that the entire game = the long run. 12 minutes is certainly NOT the long run. What the stats say, however, is if you are making decisions that are risky, but very profitable in theory, you will likely lose. ------ The overall point is, the game needs to be structured in such a way that the sum of the decisions throughout the game is reflected in the final result. Right now, it appears to have a model where that is less likely, and rather the decisions in the beginning of the game have more weight than those later. The problem is, this WILL NOT HAPPEN. Why? Because THAT DOESNT SELL. Look at the games that have become the most popular. They are all farming games. Farming/grinding to get a comparitive advantage over an enemy. CS, WoW, CoD(rank farming), LoL, Farmville. ALL the most popular games these days are about farming. And farming is in all its implementations snowbally: the winning farmer becomes better than the losing farmer. Why is this always made snowbally? because non-snowbally farming IS NOT FUN. There is little motivation to win(get a kill, etc) if you know your opponent will gain just as much from it as you do. In short, If a game has farming in it(as popular modern games do), it will be snowbally. Thats all there is to it. You would need to have like an ARAM with gold only from timer to make it linear
You listed a bunch of games where 1 death from time to time in a PVP enviroment doesn't affect the next encounter. In WoW and CoD (I have never played farmville) your advantage happens before it all starts, this is basically like Runes/Levels/Character unlocking in LoL.
|
On June 14 2012 12:33 Kronen wrote: Perhaps we are misunderstanding each other. I'm treating snowballing as the mechanism that causes comebacks to be all but non-existent in LoL. Your saying "Snowballing is the only reason there is any value in doing anything for gold," makes me thinkg you're mistaking snowballing for just general good intelligent play. But let me clarify... I'm here because I find LoL too predictable and I wanted to see if there's math that can back up (or squash undeniably) my feelings. Having a game in which it is nigh on impossible to come back is a problem.
I'm not here to speculate on what causes the problem or how to fix it. I'm just trying to benchmark games and see if the problem exists and how it develops.
I understand snowballing to refer to the fact that once you have an advantage, it becomes (substantially) easier for you to gain future advantages, and this in a way that is explicitly intended by the game design. I believe this is the way the word is generally used by this community and in this context. (The examples about soccer players not being able to use points to buy faster shoes is in the same vein.)
Like a small snowball that may end up gradually growing into an avalanche, the term is used to describe that you can exploit a small advantage to continuously increase the advantage you have. Top lane is a typical example of this: In many matchups that are fairly even, giving up first blood to your top lane opponent means you have very little chance of regaining control of your lane without help from your jungler. AD carries are the most extreme example: feed the AD carry and you lose, because AD carries build AD, AS, and crit, which all scale multiplicatively, so they snowball (i.e. scale with gold) very hard.
Gold is the obvious reason for snowballing, but to a lesser degree the experience system causes the same (XP advantage in lane helps you win more fights). But XP caps out much sooner than gold/items, and it's much harder to deny XP than to deny last hits. And they actually changed XP gains from kills in a way that counteracts snowballing.
Whether comebacks or not are easy is not is an entirely separate question. Comebacks are easy if you allow a team to hold on to advantages by playing very conservatively, regardless of whether snowballing is allowed or not. (If snowballing happened extremely easily, comebacks would be very hard, but a ~10% gold advantage by itself really doesn't make your team that much stronger in terms of raw statistics.)
Yango pointed out ways which make it very easy to play conservatively in LoL, and that sounds like a much more likely culprit to me. If you can hold on to a small but constant advantage, you don't need any snowballing to happen, you just ride the game out carefully. That is also what people seem to lament -- that games are long and boring once the initial advantage has been secured by a team. If the game makes it easy to avoid confrontations/ganks/etc., you offer the more skilled team fewer opportunities to actually apply their skill advantage.
|
On June 15 2012 08:20 cLutZ wrote:Show nested quote +On June 15 2012 07:38 brolaf wrote:On June 15 2012 05:16 cLutZ wrote:On June 14 2012 17:44 BlueSpace wrote:On June 13 2012 02:03 cLutZ wrote:On June 12 2012 22:45 BlueSpace wrote: 1) Cause and effect:
The OP has presented data which shows that a 10% gold lead by 12 minutes leads in 90% of the cases to win for the leading team. People here interpret this in a way which indicates that the rest of the game is "meaningless". You could also interpret this in way that the stronger team will manage to pull ahead early and this shows.
This is...false. Randomness decreases as gametime increases. It also decreases as the number of decisions a team, as a whole, makes. The problem? Early game is the part of the game where teams make the fewest real decisions. In reality, a few (think 3-4) are turning the outcome of the game. Particularly in pro games. Last hitting is pretty standard, so that isn't going to be a huge advantage (unless you got a kill, which is what probably happened to create this lead 1 kill, or 2), team comp could be giving you an advantage (your jungler could be faster), but usually its a bunch of RNG and highly guess-y things determining who has the lead at this point. Good examples of what I'm saying: A successful or failed invade; a successful gank, perhaps 2; a countergank, or bait bottom. Most of these are based on where the jungler is, and are very dependent on the fog of war, and getting good ward placement at this point in the game is nearly impossible. So, not only are you determining the outcome of the game during a small portion of the game, you are also determining the outcome during the period of the game that is most fundamentally random. Thats why you see so many safe picks in pro games, particularly top and mid. 1) The game start is not random at all. Everything including buffs are fixed at the game start. You have zero randomness at the beginning of the game, therefor randomness has to increase as the game progresses and cannot constantly decrease. 2) Last hitting at pro level is only standard in so far, that when left alone everybody is pretty perfect at it. In reality you have the enemy jungler and the enemy laner to worry about. The ability to last hit is not entirely decided inside the game but heavily influenced by the picks and bans prior to game start. Getting outpicked will lead to a bad lane matchup which in turn will lead to less CS as the enemy laner can harass more effectively. There is nothing random about that. 3) Jungling is not random. The jungler does not randomly run around the map and ganks lanes. He can't because he will fall behind in farm. Positioning of the jungler is usually quite well understood and there are specific timings for ganks. Pro players know when a lvl 2 gank might happen for example. The counter is to not overextend. Not Random... 4) Ward placement in the early game is not hard. Everybody is in their lane. You know from where the enemy jungler will come and all towers are up. Warding late game is much harder since you might be down several towers and at any given time several people might be missing from their lanes and lurking in the bushes, waiting to attack you. Again no randomness... I don't want to say that there is no luck involved in LoL, but there is no inherent randomness to the early game it is as skill and comp based as any other part of the game. Random was, of course, the wrong word to use. What I should have said is "uncertainty". 1. True. 2. Not relevant to this, except to the point that this reinforces my point that the snowballiness of the game significantly reduces the number of viable champions and team compositions. 3. Again, not random, uncertainty. You can't control where the jungler is, and, particularly top lane, can't have enough wards to protect itself. 4. Warding late game is easier because you have more gold and you are grouped as a team so warding is less necessary. You are basically saying that caution=skill. And in the laning phase that is almost entirely correct. Janna throwing a tornado into a bush, even if no one is there, is better off wasting both the mana and the CD than not and saving it for an aggressive play. I'm saying that this means a bunch of other things that are bad. DO you remember when M5 was super dominant with their jungle Alistar stuff? Guess what, it was based on this basic principle: get an early kill or 2 = win game. On June 14 2012 23:29 r.Evo wrote:The "randomness" you're describing doesn't exist if you look at the decisionmaking process. And that is the process which matters. The actual outcome has a very slight random element but that's not what we're talking about. The "random factor" you're talking about is plain and simple ones inability to narrow down the possibilites due to limited information. Not sure why you call me out on that when I even wrote myself that crit chance is as close as you can get to something random. As close as you can get, because it is not random. Both DotA and League of Legends use a pseudo-random distribution for crit-chance, here's an explanation of the old DotA one, not sure what changed lately: http://www.dotastrategy.com/forum/ftopic18287.html ... LoL uses a similar, modified, system but it's also not really random. The difference is, a smart player can narrow down the possible outcomes of a random/unknown event (in poker, think of pinning your opponent on a particular range of hands) and often prepare for or at least EXPECT the remaining possible [random or simply unknown] possibilities. What you're describing is not about random events. The good player understands the possibilites that are out there, narrows the opponents behaviour down to ranges and then makes a profitable decision based on those ranges. That eliminates the random factor in the long run completely. Yes, completely.Sure, it can happen that a certain player makes a huge misclick which ends up with him in a clever spot no one involved thought about before which then leads to an unaccounted kill. But that isn't about randomness in the design of the game.An actual example for randomness via design would be Chaos Knight from DotA. =P I underlined the important portion of your text. IN THE LONG RUN. Let us assume that the entire game = the long run. 12 minutes is certainly NOT the long run. What the stats say, however, is if you are making decisions that are risky, but very profitable in theory, you will likely lose. ------ The overall point is, the game needs to be structured in such a way that the sum of the decisions throughout the game is reflected in the final result. Right now, it appears to have a model where that is less likely, and rather the decisions in the beginning of the game have more weight than those later. The problem is, this WILL NOT HAPPEN. Why? Because THAT DOESNT SELL. Look at the games that have become the most popular. They are all farming games. Farming/grinding to get a comparitive advantage over an enemy. CS, WoW, CoD(rank farming), LoL, Farmville. ALL the most popular games these days are about farming. And farming is in all its implementations snowbally: the winning farmer becomes better than the losing farmer. Why is this always made snowbally? because non-snowbally farming IS NOT FUN. There is little motivation to win(get a kill, etc) if you know your opponent will gain just as much from it as you do. In short, If a game has farming in it(as popular modern games do), it will be snowbally. Thats all there is to it. You would need to have like an ARAM with gold only from timer to make it linear You listed a bunch of games where 1 death from time to time in a PVP enviroment doesn't affect the next encounter. In WoW and CoD (I have never played farmville) your advantage happens before it all starts, this is basically like Runes/Levels/Character unlocking in LoL. yeah but in wow what youre fighting for is rating,and the better your rating the easier it is to get, so the ultimate scoresheet is subject to snowballing dynamics
|
United States47024 Posts
On June 15 2012 07:31 brolaf wrote: This is a fundamental game design problem. If you want to avoid it, you need to have kills and towers to not give any gameplay advantage, and points instead like in sports. the whole concept of farming makes this impossible, as competitive farming is all about getting a comparitive advantage. lol would need to be made into a purely PvP game based on score then It's a matter of degrees.
Some amount of snowballing is an issue endemic to the genre. But even within the genre, comebacks are much more common and the game less snowbally in something like DotA.
There are certain factors in LoL that exacerbate this issue.
|
Has anyone else considered that the more expensive items in LoL are more cost efficient than the cheaper items?
In dota and hon, the more expensive items are MUCH less cost efficient than the cheaper items. Completing more expensive items is really only good because you only have 6 slots, and because of the active effects of the items.
I think that might have something to do with why games are more snowbally in LoL than in dota.
In other words, when a team is slightly ahead in dota, the extra gold they have is partially getting sunk into things like recipes and combining 2 items with no increase in stat/gold efficiency.
In LoL, however, the extra gold that they use to buy an item makes the gold that they previously spent more efficient, giving them a larger advantage than you might see by just comparing gold earned
|
There's a lot of stuff to wade through here, but one point I did not see addressed is that teams that are better will tend to exert their superior play by the 12 minute mark of any game. One thing to note from MLG is that there were really only like 6 teams out of 19 (?) that had an actual shot at winning it. That means you had a lot of instances where bad teams would go up against clearly better ones and get wrecked.
I think a more interesting exercise would be to just look at teams that were "close" in terms of skill. For example, only look at series that went to three games. (It's a crude way to restrict the data, but it makes sense to me.)
The point is, instead of thinking, "The teams that are up early tend to end up winning, therefore comebacks are very difficult," you can look at the same data and think, "Most of the time, the better team (ie, the one that wins) will tend to show their superior skill even in the first 12 minutes." So the low rate of comebacks is more a sign that the game has such a steep curve that the superior teams will show their superior ability very early on in the game.
Personally, I think both interpretations are valid. Better teams tend to take early advantages, and they tend to hold those advantages throughout the game in part because they're just better.
|
I cannot help but feel that the claims of 'snowballing' are inherent to this genre. There are no alternate objectives, no economy vs military vs tech, nothing other than plain old gold, which means when you have an advantage, it is always an absolute advantage instead of an overall (aggregate) advantage.
Until there are ways of sacrificing in one area to hopefully gain a larger advantage in another area, snowballing will always be a problem.
|
The new statistic is much more meaningful. 7% gold lead = almost no chance of comeback? I can't think that is healthy. If lack of comeback mechanism, so be it, but games need to be shorter in that case.
|
On June 15 2012 14:15 GeorgeForeman wrote: Personally, I think both interpretations are valid. Better teams tend to take early advantages, and they tend to hold those advantages throughout the game in part because they're just better.
So what you're saying is that everything is working as intended?!
|
The issue whether the statist ics are due to snowballing or to one team being much better than the other should be able to be mostly addressed by looking at bo3 series that went the full three games (or bo5 equivalent). This should at least mostly assure that teams are fairly evenly matched.
|
On June 15 2012 14:15 GeorgeForeman wrote: There's a lot of stuff to wade through here, but one point I did not see addressed is that teams that are better will tend to exert their superior play by the 12 minute mark of any game. One thing to note from MLG is that there were really only like 6 teams out of 19 (?) that had an actual shot at winning it. That means you had a lot of instances where bad teams would go up against clearly better ones and get wrecked.
I think a more interesting exercise would be to just look at teams that were "close" in terms of skill. For example, only look at series that went to three games. (It's a crude way to restrict the data, but it makes sense to me.)
The point is, instead of thinking, "The teams that are up early tend to end up winning, therefore comebacks are very difficult," you can look at the same data and think, "Most of the time, the better team (ie, the one that wins) will tend to show their superior skill even in the first 12 minutes." So the low rate of comebacks is more a sign that the game has such a steep curve that the superior teams will show their superior ability very early on in the game.
Personally, I think both interpretations are valid. Better teams tend to take early advantages, and they tend to hold those advantages throughout the game in part because they're just better.
The problem is that in series that go 2:1, the rule still holds true.
Even the team that only ends up getting 1 win (i.e. the worse team) will have this 12 minute advantage in this game and end up winning. That means even if the other team is better in general, if it is behind early by just a kill or two, there is barely any chance for a comeback.
A solution would be to make the cheaper early game items a lot more cost efficient than the lategame items, so even if you are behind in gold, you are not (simplified) half as strong as the opponent but instead ~75% as strong.
Another option would be to let skills scale better with XP and to reduce the AP/AD rates instead, so instead of for example 200+0.5 AP it would be 400+0.25 AP, items would still give a significant advantage but the difference would not be as big as it is now.
|
Can we all agree that Smoke of Deceit would make LoL far more exciting?
|
most arguments do not deny that the OP has a point. the temporal balance is sort out of wack where there is an overemphasis on the first 12 minutes, that a certain advantage in the early game is too hard to come back from. This is consistent with teams that are evenly balanced. That when one establishes and early advantage, there are no comeback mechanisms.
This means most games are decided in laning phase instead of team fights where it should be decided according to most people.
|
|
|
|