Please register for Total War Access to use the forums. If you're an existing user, your forum details will be merged with Total War Access if you register with the same email or username. For more information please read our FAQ’s here.
Notes: Below the top row of faction designations are “1,2,U”, which are the headings for Win when picking 1st, win when picking 2nd, and win when it is unknown who picked first. Look for a factions wins across the rows, with “T” being their total wins, and look for their losses down the columns, with Tot being their total losses.
For example: You can see that Beastmen have 18 wins and 17 losses, 35 total games, 51% Win rate, 75% Win Rate when picking first, 44% win rate when picking second, and 44% win rate when the pick order is unkown. Furthermore, you can see that Beastmen (BM) have 4 wins against Empire, 1 when picking first/2 when picking second/1 when the order was Unknown, and 6 losses versus Empire (EM), 2 when Empire picks first and 4 when the pick order is unkown.
This is the pick/ban tournament format data set:
Here is the corresponding Pick/Ban Tournaments used for the data set:
If you notice any tournaments that are not included that should be, please let me know!
Live streams will be pretty off by the preferences of the streamer since the skill level of the opponents on average will be much lower than that of the streamer. I think you need to separate them after who plays the faction, streamer or opponent.
Don't fear the knockdown. Control it. Embrace it. Love it!
Live streams will be pretty off by the preferences of the streamer since the skill level of the opponents on average will be much lower than that of the streamer. I think you need to separate them after who plays the faction, streamer or opponent.
Yah, this data set is already the wonkiest.... it's just for interest sake due to the exact reasons you note. My hope is that once it hits around 1000 games it will start to clean up.... if we get close to the next patch and the data is leveling out, I may go grab a few hundred casts for factions not mained by the streamers to even out the pick rates.... still not sure. At this point I'm just curious what it will look like over a cycle and want to see if it gives up anything interesting before manipulating the inputs too much.
The best thing for this data set would be to find enough streamers to have each faction represented by streamers who main them.
1 More Tournament added to Blind Pick data set.... (Paw Cup Oct 18)
Some note on how to read the spreadsheet:
Read across for your factions wins, and down for their losses. For example, the Beastmen (BM) have 23 wins and 12 losses for a 66% win rate. More specifically they have 2 wins and 3 loss vs Greenskins (GS)
This is the Blind Pick tournament format data set:
Here is the corresponding Blind Pick Tournaments used for the data set:
If you notice any tournaments that are not included that should be, please let me know!
This is stellar work man, even if I don't fully understand it. Do you think you could label some of your columns with more descriptive names to make it more readable?
Also in case you missed it, at the end of the very first post in the thread there is a video that explains the spreadsheets as well.... I know not everyone can navigate technical explanations in English so I still need more clarity on the written portion of the spreadsheets, but the walk through exists fyi.
Quite in line with my experience, skaven and liz r simply farm mode. Especially skaven tbh, those jezz r Epic noob Trap, every ratman u face u pray they bring jezz for autowin.
This sort of data shouldve been release by CA tbh
Looks like ur adding quickmatch as well? On stream or something
Quite in line with my experience, skaven and liz r simply farm mode. Especially skaven tbh, those jezz r Epic noob Trap, every ratman u face u pray they bring jezz for autowin.
This sort of data shouldve been release by CA tbh
Looks like ur adding quickmatch as well? On stream or something
Quite in line with my experience, skaven and liz r simply farm mode. Especially skaven tbh, those jezz r Epic noob Trap, every ratman u face u pray they bring jezz for autowin.
This sort of data shouldve been release by CA tbh
Looks like ur adding quickmatch as well? On stream or something
You bet! I added a third data set.... all live streams. Mostly Quick Battles, but some challenge matches & practice matches too. I'm just curious how it will compare to the other two over the cycle.
Quite in line with my experience, skaven and liz r simply farm mode. Especially skaven tbh, those jezz r Epic noob Trap, every ratman u face u pray they bring jezz for autowin.
This sort of data shouldve been release by CA tbh
Looks like ur adding quickmatch as well? On stream or something
Lotus uses fireball.........its super effective
Don't fear the knockdown. Control it. Embrace it. Love it!
You bet! I added a third data set.... all live streams. Mostly Quick Battles, but some challenge matches & practice matches too. I'm just curious how it will compare to the other two over the cycle.
Nice bro! Beautiful data collected so far, random stream or selected? Every info adds up
Nice bro! Beautiful data collected so far, random stream or selected? Every info adds up
Essentially its every stream I can find... I put out a message in general looking for suggestions and I'll add every single live stream I can source for the most part (I don't know how to sort through languages like Korean or Cantonese so a couple there are not included yet), so it is not random, its complete as far as I am able to find. The list of streams and streamers included so far are posted with the data. The point of using live streams and listing them is making my work repeatable by anyone who wants to verify it, and un-manipulated by choosing "casts" that are close/interesting/exciting, though I may do so at the end of the cycle to level out the pick rates.... I'm still not sure. At this point I'm just curious as to what it looks like, so there is no point or hypothesis being pursued on that data.
It will be recorded as a pick and ban, though as far as I can tell only the finals and battle for 3rd place were true pick and ban, while a good portion were "gentleman's agreements". I will record them under the "Unknown Who Pick First (U)" column as it is a best fit that way compared to a blind pick format for example.
It will be recorded as a pick and ban, though as far as I can tell only the finals and battle for 3rd place were true pick and ban, while a good portion were "gentleman's agreements". I will record them under the "Unknown Who Pick First (U)" column as it is a best fit that way compared to a blind pick format for example.
2 More Tournaments added to pick ban data set including the Fall Everchosen
Some notes on how to read the spreadsheet:
Notes: Below the top row of faction designations are “1,2,U”, which are the headings for Win when picking 1st, win when picking 2nd, and win when it is unknown who picked first. Look for a factions wins across the rows, with “T” being their total wins, and look for their losses down the columns, with Tot being their total losses.
For example: You can see that Beastmen have 18 wins and 20 losses, 38 total games, 47% Win rate, 75% Win Rate when picking first, 44% win rate when picking second, and 38% win rate when the pick order is unkown. Furthermore, you can see that Beastmen (BM) have 4 wins against Empire, 1 when picking first/2 when picking second/1 when the order was Unknown, and 6 losses versus Empire (EM), 2 when Empire picks first and 4 when the pick order is unkown.
This is the pick/ban tournament format data set:
Here is the corresponding Pick/Ban Tournaments used for the data set:
If you notice any tournaments that are not included that should be, please let me know!
2 More Tournaments added to pick ban data set including The Polish Championship round 2 and Shetland Apache's KoTH.
Some notes on how to read the spreadsheet:
Notes: Below the top row of faction designations are “1,2,U”, which are the headings for Win when picking 1st, win when picking 2nd, and win when it is unknown who picked first. Look for a factions wins across the rows, with “T” being their total wins, and look for their losses down the columns, with Tot being their total losses.
For example: You can see that Beastmen have 19 wins and 24 losses, 43 total games, 44% Win rate, 75% Win Rate when picking first, 36% win rate when picking second, and 38% win rate when the pick order is unkown. Furthermore, you can see that Beastmen (BM) have 4 wins against Empire, 1 when picking first/2 when picking second/1 when the order was Unknown, and 7 losses versus Empire (EM), 2 when Empire picks first and 5 when the pick order is unkown.
This is the pick/ban tournament format data set:
Here is the corresponding Pick/Ban Tournaments used for the data set:
If you notice any tournaments that are not included, please let me know!
For those not super familiar with statistics... By the end of this weekend we should pass or be close to having a usable data set on all three spreadsheets, though they will still have fairly high variability.
What do I mean by this? When doing a 'quick and dirty" analysis, you can often get away with 3 data points to just "see where you are at". Each data point in our data set is a faction versus faction result..... so for example, Chaos vs Skaven would have to occur 3 times for us to have this "quick and dirty" feel.... though as you can quickly understand it is still highly variable.
At 3 data points per match up that gives 42 games played per faction or 630 total data points required for our first "glimpse" into the numbers. Even if a faction has not played it still gives up some information due to the absence of data.
The next important positions are really 5, 6, and 7 data points, but we may not get there by the end of the cycle if we are going to see DLC by mid December.
Request:
Because we may not make that magical 7 data points per match up (average) by mid December, I would ask that you reply in this thread with any tournaments you see, YouTubers or Twitch Streams you know of.... I do have the list of each Streamer I am using above, but If we get repeats its ok.... I just want to make sure we cross the finish line before the next patch!! I just don't want to miss data that is out there, and I'm sure there are tournaments and streamers I have not come across yet!!
What's most interesting to me here is the huge disparity between blind pick and pick-and-ban when it comes to factional win/loss ratio.
Take Beastmen, for example; in pick-and-ban they're underwater, 19/24. Pretty bad results. But in blind pick they're *hugely* superior, with a whopping 23/12, which is the best win/loss ratio of any faction.
I'm not sure what that really indicates; my first inclination would be to say that this is evidence for the Beastmen having very polarized matchups, with many 'good' matchups where they have distinct advantages but then also a few 'bad' matchups in which they're very simple to counter - so if you're blind picking then running into Beastmen is a nasty surprise most of the time, but if you're in pick-and-ban you just ban Beastmen anytime you're playing a faction that struggles with them.
Or, another possibility is that the playstyle that playing Beastmen encourages is one that's strong initially but 'brittle' and easy to counter, so players who aren't used to them get wrecked but players who know a lot about them can counter them very easily. Food for thought, in any case!
What's most interesting to me here is the huge disparity between blind pick and pick-and-ban when it comes to factional win/loss ratio.
Take Beastmen, for example; in pick-and-ban they're underwater, 19/24. Pretty bad results. But in blind pick they're *hugely* superior, with a whopping 23/12, which is the best win/loss ratio of any faction.
I'm not sure what that really indicates; my first inclination would be to say that this is evidence for the Beastmen having very polarized matchups, with many 'good' matchups where they have distinct advantages but then also a few 'bad' matchups in which they're very simple to counter - so if you're blind picking then running into Beastmen is a nasty surprise most of the time, but if you're in pick-and-ban you just ban Beastmen anytime you're playing a faction that struggles with them.
Or, another possibility is that the playstyle that playing Beastmen encourages is one that's strong initially but 'brittle' and easy to counter, so players who aren't used to them get wrecked but players who know a lot about them can counter them very easily. Food for thought, in any case!
i'm a little confused by all the charts to do it myself, but you could answer your own question in large part by seeing which matchups they win more and less in. Presumably those trends should be similar across tournament types, but the actual matchups they face will be different across tournament types.
How many charts are there with the latest results? Any chance you could update the original post with the latest data so that’s the only thing we refer to? You may have already I’m just confused...
How many charts are there with the latest results? Any chance you could update the original post with the latest data so that’s the only thing we refer to? You may have already I’m just confused...
I have been updating continuously as I go, so the oldest data is the first post, then newest data is the last post, I also keep the date on the bottom as well for reference.... I figured some people might want to see the data evolve over time. I will also update the first post as well so the newest information, so it is easier to reference everything in a single post.
2 More Tournaments added to Blind Pick data set.... Claw Cup 11 & Princes Pittance (Oct 26)
Some notes on how to read the spreadsheet:
Read across for your factions wins, and down for their losses. For example, the Beastmen (BM) have 29 wins and 17 losses for a 63% win rate. More specifically they have 3 wins and 3 losses vs Greenskins (GS)
This is the Blind Pick tournament format data set:
Here is the corresponding Blind Pick Tournaments used for the data set:
If you notice any tournaments that are not included that should be, please let me know!
Just because its easy enough to do, here are all games recorded since the Sept 11 patch (QB/Blind Pick Tournies/PickBan Tournies) added into a single spreadsheet.
Comments
Some note on how to read the spreadsheet:
Notes: Below the top row of faction designations are “1,2,U”, which are the headings for Win when picking 1st, win when picking 2nd, and win when it is unknown who picked first. Look for a factions wins across the rows, with “T” being their total wins, and look for their losses down the columns, with Tot being their total losses.
For example: You can see that Beastmen have 18 wins and 17 losses, 35 total games, 51% Win rate, 75% Win Rate when picking first, 44% win rate when picking second, and 44% win rate when the pick order is unkown. Furthermore, you can see that Beastmen (BM) have 4 wins against Empire, 1 when picking first/2 when picking second/1 when the order was Unknown, and 6 losses versus Empire (EM), 2 when Empire picks first and 4 when the pick order is unkown.
This is the pick/ban tournament format data set:
Here is the corresponding Pick/Ban Tournaments used for the data set:
If you notice any tournaments that are not included that should be, please let me know!
- Report
0 · Disagree Agree- Report
1 · Disagree Agree- Report
1 · Disagree AgreeThe best thing for this data set would be to find enough streamers to have each faction represented by streamers who main them.
- Report
1 · Disagree AgreeSome note on how to read the spreadsheet:
Read across for your factions wins, and down for their losses. For example, the Beastmen (BM) have 23 wins and 12 losses for a 66% win rate. More specifically they have 2 wins and 3 loss vs Greenskins (GS)
This is the Blind Pick tournament format data set:
Here is the corresponding Blind Pick Tournaments used for the data set:
If you notice any tournaments that are not included that should be, please let me know!
- Report
0 · Disagree Agree- Report
0 · Disagree AgreeThis sort of data shouldve been release by CA tbh
Looks like ur adding quickmatch as well? On stream or something
Top #3 Leaderboard on Warhammer Totalwar.
- Report
0 · Disagree Agree- Report
4 · Disagree Agree- Report
0 · Disagree Agree- Report
2 · Disagree Agree- Report
1 · Disagree AgreeTop #3 Leaderboard on Warhammer Totalwar.
- Report
1 · Disagree Agree- Report
2 · Disagree AgreeThanks Mods!
Thanks Tzu!
- Report
5 · Disagree Agree- Report
1 · Disagree Agree- Report
2 · Disagree Agree- Report
0 · Disagree AgreeSome notes on how to read the spreadsheet:
Notes: Below the top row of faction designations are “1,2,U”, which are the headings for Win when picking 1st, win when picking 2nd, and win when it is unknown who picked first. Look for a factions wins across the rows, with “T” being their total wins, and look for their losses down the columns, with Tot being their total losses.
For example: You can see that Beastmen have 18 wins and 20 losses, 38 total games, 47% Win rate, 75% Win Rate when picking first, 44% win rate when picking second, and 38% win rate when the pick order is unkown. Furthermore, you can see that Beastmen (BM) have 4 wins against Empire, 1 when picking first/2 when picking second/1 when the order was Unknown, and 6 losses versus Empire (EM), 2 when Empire picks first and 4 when the pick order is unkown.
This is the pick/ban tournament format data set:
Here is the corresponding Pick/Ban Tournaments used for the data set:
If you notice any tournaments that are not included that should be, please let me know!
- Report
0 · Disagree AgreeStill need some more live streamers so if you see some not on the list please let me know!!
Quick Battle Data Set:
Here is the corresponding QB Casters used for the data set:
- Report
3 · Disagree AgreeSome notes on how to read the spreadsheet:
Notes: Below the top row of faction designations are “1,2,U”, which are the headings for Win when picking 1st, win when picking 2nd, and win when it is unknown who picked first. Look for a factions wins across the rows, with “T” being their total wins, and look for their losses down the columns, with Tot being their total losses.
For example: You can see that Beastmen have 19 wins and 24 losses, 43 total games, 44% Win rate, 75% Win Rate when picking first, 36% win rate when picking second, and 38% win rate when the pick order is unkown. Furthermore, you can see that Beastmen (BM) have 4 wins against Empire, 1 when picking first/2 when picking second/1 when the order was Unknown, and 7 losses versus Empire (EM), 2 when Empire picks first and 5 when the pick order is unkown.
This is the pick/ban tournament format data set:
Here is the corresponding Pick/Ban Tournaments used for the data set:
If you notice any tournaments that are not included, please let me know!
- Report
6 · Disagree Agree- Report
0 · Disagree Agree- Report
1 · Disagree AgreeUpdate:
For those not super familiar with statistics... By the end of this weekend we should pass or be close to having a usable data set on all three spreadsheets, though they will still have fairly high variability.
What do I mean by this? When doing a 'quick and dirty" analysis, you can often get away with 3 data points to just "see where you are at". Each data point in our data set is a faction versus faction result..... so for example, Chaos vs Skaven would have to occur 3 times for us to have this "quick and dirty" feel.... though as you can quickly understand it is still highly variable.
At 3 data points per match up that gives 42 games played per faction or 630 total data points required for our first "glimpse" into the numbers. Even if a faction has not played it still gives up some information due to the absence of data.
The next important positions are really 5, 6, and 7 data points, but we may not get there by the end of the cycle if we are going to see DLC by mid December.
Request:
Because we may not make that magical 7 data points per match up (average) by mid December, I would ask that you reply in this thread with any tournaments you see, YouTubers or Twitch Streams you know of.... I do have the list of each Streamer I am using above, but If we get repeats its ok.... I just want to make sure we cross the finish line before the next patch!! I just don't want to miss data that is out there, and I'm sure there are tournaments and streamers I have not come across yet!!
- Report
1 · Disagree AgreeTake Beastmen, for example; in pick-and-ban they're underwater, 19/24. Pretty bad results. But in blind pick they're *hugely* superior, with a whopping 23/12, which is the best win/loss ratio of any faction.
I'm not sure what that really indicates; my first inclination would be to say that this is evidence for the Beastmen having very polarized matchups, with many 'good' matchups where they have distinct advantages but then also a few 'bad' matchups in which they're very simple to counter - so if you're blind picking then running into Beastmen is a nasty surprise most of the time, but if you're in pick-and-ban you just ban Beastmen anytime you're playing a faction that struggles with them.
Or, another possibility is that the playstyle that playing Beastmen encourages is one that's strong initially but 'brittle' and easy to counter, so players who aren't used to them get wrecked but players who know a lot about them can counter them very easily. Food for thought, in any case!
- Report
2 · Disagree Agree- Report
1 · Disagree AgreeAdded in a bit of Spanish Tourney data as well as I could not tell the tourny type so I dumped the data here.
Quick Battle Data Set:
Here is the corresponding QB Casters used for the data set:
- Report
1 · Disagree Agree- Report
0 · Disagree AgreeThere are 3 different data sets:
1) Pick/Ban Tournaments
2) Blind Pick Tournaments
3) Live Stream Quick Battles
There is about to be a 4th which is the sum of all battles tracked so far.
Please note.... all the data so far represents a small sample size so it has very high variance.
So, the most recent post is the latest data. I will also start updating the very first post in the thread with the latest data.
- Report
1 · Disagree AgreeSome notes on how to read the spreadsheet:
Read across for your factions wins, and down for their losses. For example, the Beastmen (BM) have 29 wins and 17 losses for a 63% win rate. More specifically they have 3 wins and 3 losses vs Greenskins (GS)
This is the Blind Pick tournament format data set:
Here is the corresponding Blind Pick Tournaments used for the data set:
If you notice any tournaments that are not included that should be, please let me know!
- Report
1 · Disagree Agree- Report
3 · Disagree Agree