EPT- added wins per matchup and over time in absolute figures (labelled "+-") instead of just %-winrates as suggested.
- added version for office 97-2003 (got twice as large though, i don't know why...). Download links below!
Hey TL,
With the start of Season 3, I began to keep track of some basic stats on my 1v1 games and I am very happy with what I got out of it so far. Therefore, I want to share the file and hope, that some of you might find it helpful as well.
Let me state, that it is specifically designed to my Zergie– needs. Terran&Protoss buddies won’t get anything out of it, I’m sorry!
The spreadsheet I created actually motivates me to ladder much more: I managed to burn all my bonus pool in Season 3 for the first time (which was my goal) and this Season I shoot for 100 Wins and I’m on track so far. Nothing special for some of you probably, but for me, it’s huge, as I only have so much time besides work & family and my feelings about the “scary” 1v1 ladder were pretty mixed most of the time before I started with the spreadsheet.
Basically, I was frustrated about my development as a player. Not only that I didn’t improve as fast, as I would have wished, but also, that I actually didn’t have many measures to tell if I actually have improved. I wasn’t happy to measure my improvement only on my rank and the opponents I get matched against, as you win about 50% of the time anyways. Especially, as this doesn’t help me at all in figuring out, what I need to do to become better, which always left an empty, rather demotivating feeling.
To my mind, there is literally hundreds of different skills or pieces of knowledge, that one can incorporate into his game in order to become better. However, it is very easy to get overwhelmed by that sheer amount (especially for lower level players) and forget to properly prioritize the right pieces to work on right now. This might, for example, lead to players thinking far too much about build orders & units compositions instead of just how to get more stuff and win anyways and then feel frustrated because their oh-so-well thought out compositions don’t work. I was guilty as charged in this regard as well!
So I created this little tool, which allows me to analyze my play “by the numbers”. And by that, I mean by the important numbers! I enter some data from the game (Spending Quotient!) and SC2gears after each match and it creates some nice graphs afterwards. So improvement becomes instantly visible and I can also see, where my focus should mostly be at the moment. I strongly believe concentrating on one specific thing to improve is the way to go.
Here’s what goes into the spreadsheet:
Mandatory Information
1) Opponents Race
2) Win or Loss
3) Ladder or Custom Game
4) Spawning ratio
5) Injection Gap
6) Creep tumors laid
7) Average Income
8) Average unspent resources
9) APM
10) EAPM
11) Game Duration
Besides that, I have some optional columns, some of which I fill regurlaly, some of them only occasionally. I generally use everything up to “Opponents build” and “What (earliest thing) won or lost me the game”. Everything else, I fill when I feel like in varying detail. This allows me, while focusing on the really basic stuff, to still satisfy the urge to go into the more complex thoughts on my gameplay, as well as to make notes on the very basic things over a few dozens of games in order to find some reoccurring problems I have.
Here’s the list of the optional columns within the file.
Optional Information
- Date
- Replay Name
- Map
- Position on map
- General Game Plan
- Opponents Build
- Reasons for Injection delays
- Supply Blocked @ x Food
- When did money shoot above 500 for the first time when not saving up for sth. or being supply blocked
- What won / lost me the game (earliest / most important thing)
- Remarks on Build Order / Executing the plan / Decision making
- What went wrong in engagements
- Scouting info (espec. Missed / misinterpreted scouting info)
Please note, that the whole process takes anything from 2 – 5 minutes only depending on the amount of verbal information I want to put in!
I would like to present to you the graphs coming from the mandatory information now with my current stats as an example.
From the mandatory information, I can calculate quite a lot of averages and immediately see, how I actually did in the last matches compared to my average. I can also see the values for each item in relation to the game duration, which can be quite helpful as well.
Let’s start with the most important thing for us Zergies…vomitting larvae!
![[image loading]](http://i.imgur.com/6s9Ps.jpg)
This graph shows the average injection gap for each of my games (blue line) and the overall average (red line). As you can see, the average is decreasing steadily over time and the amplitudes shrink towards the end. I can firmly state: that is a result of focus over 20-30 games, when I looked in into each replay, looked from my POV what I was doing, when I was not injecting and wrote it down in the respective column within the optional information. It turned out, I was doing quite a lot of things while NOT-injecting. Microing units, especially Mutas, was a big reason for delays. Damn, did I love to watch those guys fly! Also, I might have been spreading overlords / creep, hotkeying units / tech structures, building structures andnot only on a few occasions - howe mbarassing to watch – even producing units without injecting afterwards andjumping right back into the battle. Having this big list was great, because I could set some rules for myself:
- When in battle, regurlaly tap back to the base and check if I am injecting / briefly shift-move the Mutas out and shift-attack them back in if I am too afraid too loose them
- If there is a lot of larvae available, directly inject after production
- Do not build structures / hotkey units / spread overlords unless you made sure you are currently injecting
- Spread creep after vomiting larvae and producing units (but that one came later, when focusing a bit more on creep)
At the moment, improving on larvae injects is still on top of my agenda although my improvement is not flattening out to much. It is just THE thing to work on, until it is close to perfection! In order to further investigate this, I have three separate graphs, which only show the values of the last 100, 50 &25 games. As the average over all games moves rather slow, I have these more current benchmarks to measure my actual performance against.
![[image loading]](http://i.imgur.com/ej7EL.jpg)
The additional horizontal black line shows the average over all games, so I can see, how the current average of the last 100/50/25 games fares against that as well.
Besides that, there are 4 graphs that show each game’s injection gap with the resp. duration of the game. There is one chart for all games played, one for the last 100 games compared to the 100 games before that and the same for the last 50 and 25 games.
![[image loading]](http://i.imgur.com/Mztzj.jpg)
This particularly shows, that I got quite good when it comes to shorter games and became better, but not good enough yet, when it comes to longer games. So the longer a game goes, the more I try to remind myself of injecting now!
I have the very same structure set up for Spawning Ratio, Spending Quotient, EAPM and APM.
I am not 100% sure, wether Injection Gap or Spawning Ratio is the more important thing to look at (your thoughts appreciated!), but looking at this graph, which shows each game’s injection gap and spawning ratio, it occurs, that there are a couple of games to the bottom left of the heavily populated area, where I have a quite low spawning ratio with quite a short injection gap.
![[image loading]](http://i.imgur.com/6qpqa.jpg)
On the other hand, there are not really games, where the opposite is true. Therefore, I prefer injection gap for the moment.
During Season 3, Whatthefat came up with the concept of the spending quotient and – despite some flaws – what a great concept that is! Yes, you can save up and thus lowering SQ, while having a maxed army to insta-remax and so on, but at the end, your average over all games will be somewhat telling about your macro! For me, it is a side indicator right now. My main focus is still the injection gap, which, as production immediately follows afterwards, also influences my spending quotient. However, the development in this area is slower, than in injection gap. Of course there are other things to a high spending quotient: having more hatcheries and thus more units to be produced, investing in tech, not getting supply blocked and so on. So I feel like once my injection gap improvement is becoming to flat, I should focus on all these areas than. Right now, I of course think about when to get extra hatcheries and all of that, but only besides becoming better at injecting! Here are my current graphs for the spending quotient:
![[image loading]](http://i.imgur.com/Fsq4E.jpg)
![[image loading]](http://i.imgur.com/byPb1.jpg)
As you see, there is improvement, it’s just slower.
I believe EAPM and APM are controversial topics on this forum. I often read, that APM is not important at all and stuff, but I tend to disagree. I mean, in a REAL-TIME strategy game, how can it not be advantageous to play faster, if the faster does not mean just spamming mindlessly?Also, it gives me more confidence, to know, that I can execute quite a bitfaster than 200 games ago. Therefore, I really like to look at this one as well! As the graphs have the same structure as the previous ones, I won’t show them here, but you will find them in the tool.
Lets move on to the slimy stuff: Creep – Spread!!
After having made initial adjustments to larvae injects, I thought of creep – spread as an important thing to look into next. Having the earlier vision of enemy units and more abilities to flank (or even connect at all with banelings) is too big to ignore for too long. So here are those graphs:
![[image loading]](http://i.imgur.com/iVPlg.jpg)
I have some troubles with that though, as I don’t spread creep at all in ZvZ and in ZvP&ZvT it is much dependent on what I am going for in a game. If I plan to play a macro game, I usually get a third queen shortly after the second one and use it to spread creep. If I go for an aggressive build though, there is just no place for a third queen and creep spread. So the values and averages for creep spread are not really accurate, but at least let me see how I fare in each game and judge, if I did somewhat of a proper job in that regard. With that in mind, I only show the creep tumors spread in total and per minute (and an average against Terran and Protoss only), as well as how I did in the last 25 games.
I also added win rate graphs per matchup and win rates over intervals of game duration per matchup. I only graph my ladder games in these sections though, as practise games are against opponents of to varying skill, compared to my then current skill, to have a real meaning. Also, those graphs are not really nice looking, as I have a couple of problems selecting the relevant data properly. For instance, if game number 100 was a game against Terran, Zerg and Protoss will show the same win rate as in game 99, as only the Terran winrate changes with game 100. Also, in this screenshot all games are shown on the X-axis still, but only the winrate for ladder games is graphed. This means, that if a game is a custom game, all winrates stay the same for that game. You will recognize therefore, that I played relatively more custom games in the past then nowadays, because the graphs are oftentimes horizontal in the beginning.
![[image loading]](http://i.imgur.com/kTd84.jpg)
EDIT: (Gesamt = Overall, fixed in the file)
However, there is a solution in filtering the data in the data-entry spreadsheet. If you set game type to ladder only, the graph already looks better.
![[image loading]](http://i.imgur.com/K9I8H.jpg)
And if you combine this with a filter for opponents race, you will see a graph for that matchup, as it is supposed to look like:
![[image loading]](http://i.imgur.com/Vs1iG.jpg)
Yeah…my game vs. terran sucks lately…
The same flaws of selecting data and the methods of filtering apply to my time graphs. I find these graphs particularly interesting though, as it tells me a lot about the most basic strategic zerg problem: How greedy can I be?
![[image loading]](http://i.imgur.com/4fLE6.jpg)
I selected the intervals of <9 Minutes, between 9 and 14 Minutes, 14 to under 20 Minutes, 20 to 24 minutes and above 24 minutes as intervals to be graphed. You see, that I heavily sucked in the period from 9 to under 14 minutes in the past. If a game ended then, I was pretty sure to loose. However, I was pretty strong from the 14 to 24 minute mark back then. I decided to go for more aggressive styles lately and you can clearly see that in the graph as well: Now I am doing pretty well in the early midgame but became weaker afterwards. I don’t know, if that is good or bad, but I will continue to look at it in the future to find the right amount of greediness for all matchups and am sure, this graph will help me with that. Of course, this is also available per matchup.
So that’s it: I hope some of you find it useful in order to focus their improvement better than before or even feel motivated to ladder more!!
Please let me know if you can think of anything, that could or should be incorporated! Also, your thoughts on the proper time intervals would be highly appreciated!
GL HF!!!
Office 2007 & up (no macros) - Password B6gkkk3
Office 97-2003 (no macros as well) - Password zD8kkk3
Finally some instructions for data entry:
+ Show Spoiler +
First of all, I left my current data in the file. If you want to start it on your own, just delete everything in Columns A to P fom Row 2 to 256.
There are some things that need to be followed if the spreadsheet should work properly. I will run through the mandatory (green) columns from left to right and explain each of those.
1) Game number: Consecutive number of the game starting with 1 up to the current game. This is needed to calculate averages. Do not enter future numbers preemptively, because that will screw up the formatting of graphs you want to look at
2) Game Type: Enter “L” for Ladder games and whatever else you like for custom games. “L” is needed to identify ladder games
3) Opponents Race: Enter “T”, “Z” or “P” only
4) Result: Enter “W” or “L”
5) Game Minutes: Full Minutes of game duration, no decimal figures
6) Game Seconds: Full Seconds of Game duration
7) Avrg. Unspent ressources: comes from the match result page in game under the ressources tab
8) Avrg. Income: See 7)
9) Spawning Ratio: comes from SC2gears on the Tab “Main Production”
10) Average Injection Gap: See 9)
11) Creep Tumors: Comes from SC2 gears. Just go to the Tab "Builds/Tech Stat" and you will see all items you build the entire game, creep tumors amongst them (Thx Dakota_Fanning!"
12) APM / EAPM come from SC2gears on the tab “APM”
13) Yellow Columns can be used without any restrictions. Red columns should not be changed, as they are calculated automatically.
14) File is restricted to 2.500 games right now, because of exploding file size. If you need more lines, just let me know and I can tell you what to do
There are some things that need to be followed if the spreadsheet should work properly. I will run through the mandatory (green) columns from left to right and explain each of those.
1) Game number: Consecutive number of the game starting with 1 up to the current game. This is needed to calculate averages. Do not enter future numbers preemptively, because that will screw up the formatting of graphs you want to look at
2) Game Type: Enter “L” for Ladder games and whatever else you like for custom games. “L” is needed to identify ladder games
3) Opponents Race: Enter “T”, “Z” or “P” only
4) Result: Enter “W” or “L”
5) Game Minutes: Full Minutes of game duration, no decimal figures
6) Game Seconds: Full Seconds of Game duration
7) Avrg. Unspent ressources: comes from the match result page in game under the ressources tab
8) Avrg. Income: See 7)
9) Spawning Ratio: comes from SC2gears on the Tab “Main Production”
10) Average Injection Gap: See 9)
11) Creep Tumors: Comes from SC2 gears. Just go to the Tab "Builds/Tech Stat" and you will see all items you build the entire game, creep tumors amongst them (Thx Dakota_Fanning!"
12) APM / EAPM come from SC2gears on the tab “APM”
13) Yellow Columns can be used without any restrictions. Red columns should not be changed, as they are calculated automatically.
14) File is restricted to 2.500 games right now, because of exploding file size. If you need more lines, just let me know and I can tell you what to do
