|
Post by zer00eyz on Jan 18, 2018 20:50:34 GMT -5
Under a certain set of (limited) conditions War Robots’s MM should lead to some fairly predictable outcomes if the MM is “balanced” for the most part. The reality is that outside the boundaries where resources are MOST important (the very bottom i.e. bronze and below, and the very top i.e. champs) things appear to be predictable. The question is HOW predictable? The answer might be frighteningly so. Recently Dr. Yat published an article that highlighted the following data-points: That first point immediately jumped out at me, because I have independent data that can verify that claim under certain circumstances. That having been said lets clean up his original statement to fit that information. A team playing 5 members who are looking to land in the top spot have a 25% chance of wining against a full group of six. A team playing 5 members 4 looking to land in the top spot and 1 classic capping hanger (lights) have a near zero chance of winning against a full group of six. (Note: this is my data and has that brings this up from zero, based mostly on map however it is NOT enough to claim even a percentage point) Because I lack data on “camping” (something I will be tracking in a future project) and our dear Dr’s data is accurate on the whole, lets take his information on campers at face value. With these two bits of information in mind (win rates for campers and missing teammates), we can build a statistical model that maps to our games outcomes. My first instinct was to build something a bit more complex that what was ultimately required. Initially I suspected that by choosing a team member at random, and then eliminating from the opposite team I would get something that resembled the actual game. Instinctually this makes more sense, because with a larger advantage of bots on the field this would seem to reflect reality. Sadly this resulted in only a 12% win rate for our short handed blue team. What got me the results I was seeking was to simply pick a red or blue team members bot to eliminate, reducing the total number of bots available one by one. This process resulted in the magical %25 win rate. Along a similar line, I slowly began to increase the total number of bots to simulate what a camper might do (dwelling in a bot too long, and not leveraging their full hanger depth) - A camper who stays in their bot the entire match, 29% win rate, who leverages 2 bots, 35% win rate, who leverages 3 bots easily rounds up to 40% win rate. Not only do these line up nicely with the Dr’s information (I will seek to independently verify this in the coming weeks, but it will require a lot of sample data) but points to the fact that we can model games on pure chance. Lastly the chronic lament found of this forum, of “MM keeps me at %50” helps support that this simple model is fairly credible - MM by design will normalize out skill vs gear (we know it works that way by design) - Based on this information is there any conclusions we can draw? - Hardcore campers are nearly as damaging to gameplay as those who don’t drop bots (or the single cossack) - Your odds of wining with a camper are 29% and with a non participating member 25% - That 4% delta is insignificant.
- 5v6 with one member who is less than 100% combat effective (capping, camping) will increase your loss rate to almost zero - statistically it should be about 8% but map and bot dependence begin to weigh too heavily for our simple model I suspect.
- Dwelling too long in your long range bot isn’t doing any one any good - get out too late and you reduce your teams combat effectiveness massively (50%)!
- Eliminating campers on the red team is NOT a smart play unless they present a threat. A zeus furry is a bigger risk on canyon than on dead city on leaving them alone can be to your benefit because their “dwell time” and “combat effectiveness” are linked. #SaveRedCampers should be the motto of the blue team.
- However, eliminating combat effective bots should be a high priority. Everyone seems to grasp that focusing on a high priority target is in the teams best interest.
- The path to victory is in the total number of combat effective bots (however it is unlikely that this is the path of highest reward, this model isn’t sufficient to predict that and the one that I had is no longer representative - I will happily build a newer more robust one).
- This model leans heavily in favor of “slots over bots” - and reflects an accurate view of win rate, but not player happiness.
Notes: - I built several randomized models, and oddly coin flipping to pick a team member to eliminate fits our known data points the best
- We have a parity issue: tankers bottom out at a %20 win rate, but victory rates in 5v6’s sit at %25 - the discrepancy might be a variant of the monte hall problem or a clue to the inner workings of MM. This needs to be explored.
- This is a random model, it fits our know data well, and would indicate that MM is performing “fairly” though probably not in a sastfiying way.
- This model does not account for exceptionalism, or happiness
- We need hard numbers on games impacted by campers vs tankers to get to an overall loss rate - I will be keeping track of these things in my next sets of games to get HARD data on both and see where it fits in. I suspect that campers are "league dependent" and may have a larger overall impact than any one suspects.
|
|
|
Post by T34 on Jan 18, 2018 22:42:30 GMT -5
I may be missing something really obvious and prepared to wipe the egg off my face but I do not support the conclusions from the above test with the evidence presented. The only factual matter that was presented by @rwhereyat is the effect of camping. All other stuff is unsubstantiated.
Re the actual camping test
The author preformed a test of 20 matches and I am not sure how he drew a conclusion that a camper will win between 30-40% because the data does not fit his conclusion. Seemingly he relied on the fact that his win/loss dropped from 50% to 44%. Even that is not between 30-40%. Maybe he extrapolated the results over 50 matches and thus the drop would be 33%? Never the less, all of those are incorrect conclusions based on the data. Rather big mistake in method.
The outcomes of the test suggest that a camper will win 50% of matches and lose the other 50%. After all, he played 20 games and won 10 and lost 10. His ratings didn’t drop because he was camping but they dropped because amongst the last 50 matches he had a win streak or something that was sunsetted as a result of playing 20 matches and a loosing streak was still counted. What was left equated to 44%. If the 10 loss 10 win trend continues over the next 30 matches he would end up with a 50% win/los. There is no way on this earth that after playing 20 games resulting in 10 wins and 10 losses a conclusion can be made that the average win loss rate is between 30-40%
So the correct conclusion should be that a camper does not have a negative effect on outcomes provided we assume that a 50% win/loss is the long term average.
All hail the campers. They have been vindicated rather than condemned by evidence.
PS: The 25% win/loss rate for having a tanker is only correct under certain circumstances for a very limited time. It is way off from being correct for long term trends. Long term trend is 38% to 40%. Without going into detail the net effect of the tanker is an average of 11% increased chance of loosing a particular game but has negligible if any affect on the long term trend. If the MM is set to 50% it has no effect by definition.
|
|
|
Post by Dejnov on Jan 18, 2018 22:47:16 GMT -5
Nice analysis; very interested in the assumption that you shouldn’t eliminate the long range bots. That’s an interesting tactic to try out. Also makes me feel better about my all brawling hangar...
Dejnov
|
|
|
Post by zer00eyz on Jan 18, 2018 22:56:53 GMT -5
PS: The 25% win/loss rate for having a tanker is only correct under certain circumstances for a very limited time. It is way off from being correct for long term trends. Long term trend is 38% to 40%. Without going into detail the net effect of the tanker is an average of 11% increased chance of loosing a particular game but has negligible if any affect on the long term trend. If the MM is set to 50% it has no effect by definition. PS: It is way off from being correct for long term trends. Long term trend is 38% to 40%. --- I would really need a data to back this claim up because it doesn't match up at all with KNOWN datapoints. 1. It doesn't align with my past (already published) and current (to be published) data sets - 25% is the average 2. Dr. Yat data aligns with 25% as well. 3. Tankers bottom out at about %20 loss rate (close enough to 25%, and far removed from your %38-40) As I stated, camping data from Dr. Yat needs a larger sample (something I will collect) but it lines up as well with the statical model (random chance) too.
|
|
|
Post by ᎶƦ℮℮ƊᎽ ƤΛƝƊΛ on Jan 18, 2018 23:21:17 GMT -5
Tankers cause win rates of 20%. That isn't an assumption or model. That's data from thousands of actual games from actual tankers. So even if campers work out to 29%, it is a difference between that and 20%, not 25%. I'm still not sure where that number comes from, as I can't get any proof of it or see any data that backs it up. Dr Yat provided no data, simply said he had data. I've never seen your data.
The camper thing would also better be figured out from an actual player camping in one bot intentionally. I understand the premise of what you're getting that number from, but the starting point is this magical 25% that a model was put together to obtain the predetermined number. You normally build a model and see what comes out and adjust variables from there. In this case, you picked an outcome, and made your model spit that out. The application of the model represents what would happen if all 30 bots faced off against all 25 bots at the same exact time, and that 100% of games are determined by which team mechs out first.
But as far as I can tell, using your assumption of campers causing 29% win rates, and tankers causing 20% win rates(based on actual data from tankers), and assuming a balanced matchmaker forces a win rate of 50%, we get the following: Campers cause a 42% deviation from the normal win rate. Tankers cause a 60% deviation from the normal win rate. Based on this, we can conclude that tankers are approximately 43% more harmful to a team's chance to win than are campers.
I know we've discussed on Discord but I feel it would be interesting to open up other opinions from the forum.
|
|
TheDokiDoki
Destrier
I'm procrastinating
Posts: 115
Karma: 74
Pilot name: TheDokiDoki
Platform: Android
Clan: Doki Doki Empire
League: Expert
Server Region: North America
Favorite robot: Hover
|
Post by TheDokiDoki on Jan 18, 2018 23:28:10 GMT -5
Too long, didn't read.
|
|
|
Post by T34 on Jan 19, 2018 0:04:21 GMT -5
PS: The 25% win/loss rate for having a tanker is only correct under certain circumstances for a very limited time. It is way off from being correct for long term trends. Long term trend is 38% to 40%. Without going into detail the net effect of the tanker is an average of 11% increased chance of loosing a particular game but has negligible if any affect on the long term trend. If the MM is set to 50% it has no effect by definition. PS: It is way off from being correct for long term trends. Long term trend is 38% to 40%. --- I would really need a data to back this claim up because it doesn't match up at all with KNOWN datapoints. 1. It doesn't align with my past (already published) and current (to be published) data sets - 25% is the average 2. Dr. Yat data aligns with 25% as well. 3. Tankers bottom out at about %20 loss rate (close enough to 25%, and far removed from your %38-40) As I stated, camping data from Dr. Yat needs a larger sample (something I will collect) but it lines up as well with the statical model (random chance) too. Few things. The data is not mine but I consider the information very reliable (and its from the live server). Same result from multiple sources. Almost identical. Happy to take it offline. Send me a PM if you like. You need to bear in mind that your outcome is valid for a particular phase of the tanking activity and a potential fluctuation within the results. What the win return is for tanking is most directly related to the win percentage when the actual tanking activity is initiated. In general, If you have a high win rate your tanking loss rate will be substantial (almost all). You have a low win rate (like 20%) your wins will be well over 50%. Deviations from the norm do happen but you can deal with them using the concept of percentiles. Its not about drwhereyat ‘s data size. He simply got the interpretation totally wrong based on the factual evidence he presented. Perhaps what he did is described the most ethical tanking method and proved that it has no affect on others in the game. How ironic ? A hard lined anti tanker presenting the best proof so far that lowering one’s league ranking is possible without affecting anyone else. Monumental! By the way, not sure how much randomness there is in this MM. This is some piece of art to keep so many people in line while also assisting with the selling effort. I mean, the bottom line is if the MM can keep most within a programmed win/loss rate this thing has to know the battle outcomes most of the time at the time of allocation. Staggering implication in relation to the illusion of competitive battle as there isn’t one. Remember some while ago when the MM was savage and induced losing and winning streaks. When I was on a loosing streak I looked around after spawning and thought to my self, poor blue guys, you are all about to lose and you don’t even know it. Ant it came to fruition.
|
|
|
Post by zer00eyz on Jan 19, 2018 0:05:44 GMT -5
Tankers cause win rates of 20%. That isn't an assumption or model. That's data from thousands of actual games from actual tankers. So even if campers work out to 29%, it is a difference between that and 20%, not 25%. I'm still not sure where that number comes from, as I can't get any proof of it or see any data that backs it up. One set of data (from the past, and theres more I can dig up) and I'll have a fresh set shortly - (from current MM) war-robots-forum.freeforums.net/thread/3133/data-70-boa-matchesBut as far as I can tell, using your assumption of campers causing 29% win rates, and tankers causing 20% win rates(based on actual data from tankers), and assuming a balanced matchmaker forces a win rate of 50%, we get the following: Campers cause a 42% deviation from the normal win rate. Tankers cause a 60% deviation from the normal win rate. Based on this, we can conclude that tankers are approximately 43% more harmful to a team's chance to win than are campers. I know we've discussed on Discord but I feel it would be interesting to open up other opinions from the forum. Your math is a bit off: If your chances of winning 5v6 are 20%, and your normal chance in any match is %50, that change IS %60 deviation. If your chances of winning 5v6 are 25%, and your normal chance in any match is %50, that change IS %50 deviation. If your chances of winning with the worst camper are 29% and your normal chance in any match is %50, that change IS %42 deviation. However, you can only really use the difference in the deviation to determine how much more impactful a tanker is vs a camper -- between 8% and 18% more harmful (depending on numbers used) - no clue where you got 43% from... --------------------------------------------------------------- Lets assume that the 25% 5v6 and 20% floor for tankers are are BOTH accurate. Lets also assume that tankers appear with NORMAL distribution (1 in every N games randomly over red and blue). The odds of winning when a 2nd tanker shows up on red go UP to 50/50 but DOWN to 89.9/10.1 when there is a 2nd tanker on blue. The math escapes me but we should be able to use the discrepancy to check the frequency of tankers. Assuming that checks out, we could look at a current data set and check it against the actual frequency of 5v6's!
|
|
|
Post by bronzeknee on Jan 19, 2018 0:07:38 GMT -5
Excellent analysis, thank you for it. A couple of questions:
Are we measuring full group of six as six players who stayed in the game the entire time? I find there are many games where I think I win a 5v6, only to find out one of their players did 50k damage then left, so it was really more like a 5vs5.
This statement seemingly contradicts this one:
Also, how do we know that 4% isn't significant? Was statistical significance calculated?
|
|
|
Post by T34 on Jan 19, 2018 0:08:39 GMT -5
Tankers cause win rates of 20%. That isn't an assumption or model. That's data from thousands of actual games from actual tankers. So even if campers work out to 29%, it is a difference between that and 20%, not 25%. I'm still not sure where that number comes from, as I can't get any proof of it or see any data that backs it up. Dr Yat provided no data, simply said he had data. I've never seen your data. The camper thing would also better be figured out from an actual player camping in one bot intentionally. I understand the premise of what you're getting that number from, but the starting point is this magical 25% that a model was put together to obtain the predetermined number. You normally build a model and see what comes out and adjust variables from there. In this case, you picked an outcome, and made your model spit that out. The application of the model represents what would happen if all 30 bots faced off against all 25 bots at the same exact time, and that 100% of games are determined by which team mechs out first. But as far as I can tell, using your assumption of campers causing 29% win rates, and tankers causing 20% win rates(based on actual data from tankers), and assuming a balanced matchmaker forces a win rate of 50%, we get the following: Campers cause a 42% deviation from the normal win rate. Tankers cause a 60% deviation from the normal win rate. Based on this, we can conclude that tankers are approximately 43% more harmful to a team's chance to win than are campers. I know we've discussed on Discord but I feel it would be interesting to open up other opinions from the forum. Panda, forget the tankers for now. look at the logic for "yat" to draw that conclusion. its wrong. he played 20 games and won 10 and lost 10. that means campers have absolutely no affect on the outcome. Yat proved it him self. ETHICAL TANKING has been vindicated by the ardent anti tanker.
|
|
|
Post by ᎶƦ℮℮ƊᎽ ƤΛƝƊΛ on Jan 19, 2018 1:05:41 GMT -5
Here is some math the anti-tanker fans will love. Tankers existence pushes your win rate up.
Here's the simple math, and we are assuming the matchmaker evenly(randomly) distributes tankers.
In every single game you play, you are a constant. Meaning every single game you experience, you are a player in the game, and you are always on blue team.
That means that in every single game you play, there are only 11 randomly assigned players. You (every single time on blue), six randomly assigned red opponents, and five randomly assigned blue teammates.
Now it gets fun, throw in a tanker in any random game. There is a 5/11 chance of the tanker randomly being assigned as one of your 5 teammates. There is a 6/11 chance of the tanker being randomly assigned as one of your 6 opponents.
This means that over time, the very existence of tankers pushes your win rate higher, because they almost 22% more likely to be on the other team (. 55/.45).
Now I understand that tankers ruin game play regardless of which side they're on. This simply debunks the theory that tankers are hurting you personally. They're actually helping your win rate. Ironic.
|
|
|
Post by hon_shu on Jan 19, 2018 1:25:26 GMT -5
And there I thought War Robots was about converting virtual robots into virtual piles of metal garbage ... Could you please elaborate a little on the model you used, zer00eyz ? And what's the input, what's the output? Then we can continue the incoherent academic bickering our friend krebby likes so much
|
|
|
Post by zer00eyz on Jan 19, 2018 1:35:10 GMT -5
Excellent analysis, thank you for it. A couple of questions: Are we measuring full group of six as six players who stayed in the game the entire time? I find there are many games where I think I win a 5v6, only to find out one of their players did 50k damage then left, so it was really more like a 5vs5. This statement seemingly contradicts this one: Also, how do we know that 4% isn't significant? Was statistical significance calculated? Based on data available (sampling of games linked elsewhere in the thread now), and coin flip player elimination (the statistical model i slapped together) you get a 25% chance of wining a 5v6. Were not accounting for "participation level" at all, that might be the reason (players get complacent when they have an advantage). > Also, how do we know that 4% isn't significant? Was statistical significance calculated? The %4 shift in odds of winning with a tanker (25%) and a zeintard (as an example of the worst kind of camper) who stays in the bot till the bitter end (29%) is "insignificant" when comparing campers to tankers. Campers suck almost as bad as tankers! The 30%-40% range that the good doctor came up with is fairly easy to model, by simply adding more bots in: the camper dies, or grows a brain, or only gets in his bot tent mid game - by playing 2 or 3 bots his odds of winning go UP in direct relationship to his stupidity decreasing... As I have stated, I'll be gathering game data from here on out to confirm if this is the case!
|
|
|
Post by U Camp = I Camp on Jan 19, 2018 1:38:48 GMT -5
Here is some math the anti-tanker fans will love. Tankers existence pushes your win rate up. Here's the simple math, and we are assuming the matchmaker evenly(randomly) distributes tankers. In every single game you play, you are a constant. Meaning every single game you experience, you are a player in the game, and you are always on blue team. That means that in every single game you play, there are only 11 randomly assigned players. You (every single time on blue), six randomly assigned red opponents, and five randomly assigned blue teammates. Now it gets fun, throw in a tanker in any random game. There is a 5/11 chance of the tanker randomly being assigned as one of your 5 teammates. There is a 6/11 chance of the tanker being randomly assigned as one of your 6 opponents. This means that over time, the very existence of tankers pushes your win rate higher, because they almost 22% more likely to be on the other team (. 55/.45). Now I understand that tankers ruin game play regardless of which side they're on. This simply debunks the theory that tankers are hurting you personally. They're actually helping your win rate. Ironic. Now you're really gonna get them going.
|
|
|
Post by zer00eyz on Jan 19, 2018 2:05:17 GMT -5
And there I thought War Robots was about converting virtual robots into virtual piles of metal garbage ... Could you please elaborate a little on the model you used, zer00eyz ? And what's the input, what's the output? Then we can continue the incoherent academic bickering our friend krebby likes so much I actually built quite a few basic models to see what would come out. The first thing I built was a simple count of bots available and then choosing one to eliminate at random (from the total available) -- the data this generated did not align with what happened in game. The output of 10k runs of 5v6 looks something like this: Redwins: 5522 Bluewins: 4478 -- that 55/44 split actually aligns nicely with T34's suggestion of how things should look -- however it doesn't reflect my data, the dr's data, and our 20% floor data from tankers (it isn't even remotely close)... My next thought was that I would need to "weight" players. TO randomly select a bot, and then to eliminate one from the opposite team. This should give an advantage to the reds in the 5v6 scenario that might shift the numbers "down" to our known game data. The outcome of this swings wildly in the other direction: Redwins: 8709 Bluewins: 1291 and again doesn't look like our known data. Thinking that "over weighing" was my issue I actually built a model of two sets redTeam = {5, 5, 5, 5, 5, 5} and blueTeam = {5, 5, 5, 5, 5} - (30 red 25 blue) and selected a team member at random to be the "winner" (so red would be favored) and then picked a team member from the OPPOSITE team at random who's set value would be decremented. The results end up looking like this: Redwins: 8853 Bluewins: 1147 (a slight reduction blue wins over the previous weighted model. What finally worked is what was described above, count the total number of bots on each team (25, and 30 respectively), and then flip a coin to choose a bot to eliminate: Redwins: 751415 Bluewins: 248585 (yes that is a million simulations) -- If we simply adding bots to this model (26,27,28 on the blue side) we can simulate the behavior of a camper and come up with numbers that match the Dr's on that front as well.
|
|
Deleted
Deleted Member
Posts: 0
Karma:
|
Post by Deleted on Jan 19, 2018 6:26:35 GMT -5
Now I understand that tankers ruin game play regardless of which side they're on. This simply debunks the theory that tankers are hurting you personally. They're actually helping your win rate. Ironic. Hah, that is funny. In seriousness, however, the "ruin game play" far outweighs any benefit from "helping your win rate".
|
|
|
Post by blueb4sunrise1 on Jan 19, 2018 6:51:31 GMT -5
This is why my heart sinks when my team start out with 4 Natashas set up for long range combat all jostling to get the best camping spot.
Don't get me wrong, I think that a sniper or bombadier is a useful addition on some maps (more so in domination mode).
I am not bashing every one that has a sniper Natasha. Its just that these long range combatants have limited potential and need to recognise that they may need to backpack around the map a bit and be prepared to eject and jump into something set-up for closer combat when required.
|
|
|
Post by MΞCHДИISΓIC on Jan 19, 2018 7:23:09 GMT -5
[quote author=" blueb4sunrise1" source="/post/243144/thread" timestamp="1516362691” Its just that these long range combatants have limited potential and need to recognise that they may need to backpack around the map a bit and be prepared to eject and jump into something a set-up for closer combat when required. [/quote] Exactly....but rare in my experience.
|
|
|
Post by ᎶƦ℮℮ƊᎽ ƤΛƝƊΛ on Jan 19, 2018 10:29:02 GMT -5
Now I understand that tankers ruin game play regardless of which side they're on. This simply debunks the theory that tankers are hurting you personally. They're actually helping your win rate. Ironic. Hah, that is funny. In seriousness, however, the "ruin game play" far outweighs any benefit from "helping your win rate".
Did I not word the paragraph well? I'm confused why you took exactly what I said and restated it.
|
|
Deleted
Deleted Member
Posts: 0
Karma:
|
Post by Deleted on Jan 19, 2018 12:04:28 GMT -5
Hah, that is funny. In seriousness, however, the "ruin game play" far outweighs any benefit from "helping your win rate".
Did I not word the paragraph well? I'm confused why you took exactly what I said and restated it. Dude I am agreeing with you and complimenting you fot seeing an amusing angle on it
|
|
|
Post by ᎶƦ℮℮ƊᎽ ƤΛƝƊΛ on Jan 19, 2018 12:22:01 GMT -5
Did I not word the paragraph well? I'm confused why you took exactly what I said and restated it. Dude I am agreeing with you and complimenting you fot seeing an amusing angle on it Oops! There's been a lot of misconstruing of my statements lately so I jumped the gun. Woke up on the wrong side of the bed today. Let's move along.. Nothing to see hear...
|
|
|
Post by mobiuswr on Jan 19, 2018 12:37:12 GMT -5
SuperHero In light of this new data. Thanks to zer00eyz . Can we please ban/lock all mentions of how to effectively Beacon Cap and Effectively Camp (use any long range weapons). We now know these behaviors are equally as detrimental to winning as Tanking. And not trying to win is against the values and principles you stand for. EDITOh yes, I almost forgot. Also please retroactively ban all members who ever endorsed using a Beacon Capping hangar or ever endorsed any long range (over 600m) camping build.
|
|
|
Post by zer00eyz on Jan 19, 2018 12:58:02 GMT -5
mobiuswrYour still under quarantine but your reading the wrong things into it. 1. I make no statements about "long range" being bad universally. I gave an example of two places where a zeus furry would require a very different response. With the advent of hanger decks building 2nd hanger with long and mid range is going to be almost a requirement. 2. You have drawn a spurious conclusion regarding capping hangers, and actually my own data from running them. The high win rate of a capping hanger (can be held steady in the 60%-%65 range) can make an outsized battlefield contribution - but NOT in a 5v6 where it is forced to brawl. With enough data (that we will never get) we could figure out the combat effectiveness of any given bot in any given situation, and I'm sure that we would find that things are "mostly" deterministic. Lastly, careful about what conclusions you draw from this data - it is talking about outcomes in a situation not how often those situations occur.
|
|
|
Post by mobiuswr on Jan 19, 2018 12:59:36 GMT -5
mobiuswr Your still under quarantine but your reading the wrong things into it. 1. I make no statements about "long range" being bad universally. I gave an example of two places where a zeus furry would require a very different response. With the advent of hanger decks building 2nd hanger with long and mid range is going to be almost a requirement. 2. You have drawn a spurious conclusion regarding capping hangers, and actually my own data from running them. The high win rate of a capping hanger (can be held steady in the 60%-%65 range) can make an outsized battlefield contribution - but NOT in a 5v6 where it is forced to brawl. With enough data (that we will never get) we could figure out the combat effectiveness of any given bot in any given situation, and I'm sure that we would find that things are "mostly" deterministic. Lastly, careful about what conclusions you draw from this data - it is talking about outcomes in a situation not how often those situations occur. Zeus-anything is Midrange, not long range. A beacon capping hangar hurts your side under the very likely circumstances you will have a tanker on your side in low leagues and is completely unviable in high leagues.
|
|
|
Post by Danny Linguini on Jan 19, 2018 13:36:07 GMT -5
|
|
|
Post by ᎶƦ℮℮ƊᎽ ƤΛƝƊΛ on Jan 19, 2018 13:41:06 GMT -5
Not even close. Not every thread that includes the word "tanker" is the same. This is an actual analysis thread using data and math. Math is this cool concept that allows you to actually prove or disprove things based on data. Not to mention, the data is clearly skewed against tankers. The point is just that it is also skewed against hardcore campers. That doesn't mean tanking is good, it means that both play styles are detrimental to gameplay (which nobody disagrees with). I actually pointed out that tankers are more detrimental than his "data" suggests. Because I have actual data as a tanker that proves tankers cause an actual win rate of 20% (and I posted that as well as a comparison to his hypothetical 29% from campers). Keep in mind all of the analysis is hypothetical, because it's based on a model that is flawed to begin with.
|
|
|
Post by zer00eyz on Jan 19, 2018 14:05:27 GMT -5
And here I thought I was safe posting data but the climate change denial conference got out early I see. (laugh at this it is a joke, but your tears are still delicious).
|
|
|
Post by spectre9 on Jan 19, 2018 14:23:43 GMT -5
How are these data being collected?
Seems like some screen captures with OCR could automate some of the work of collecting data, although it would only truly capture those with 0 damage as tankers.
|
|
|
Post by zer00eyz on Jan 19, 2018 14:33:21 GMT -5
How are these data being collected? Seems like some screen captures with OCR could automate some of the work of collecting data, although it would only truly capture those with 0 damage as tankers. Manually -- I have a fairly good screen shot workflow that is becoming second nature again - but as I'm going to start on a deeper dive that will have to be "modified" (examining player screens). It is pretty easy to go back through them and capture data into google sheets when I have the time to sit and process it rather than "play". OCR works up to a point, and I have been toying with a system to harvest data from screen shots - some player names have proved to be a blocking issue with most of the open source text parsing out there. I did stumble across a commercial product that I had a trial license with and though it was slightly better at not producing garbage however it didn't provide enough benefit to be worth pursuing.
|
|
Deleted
Deleted Member
Posts: 0
Karma:
|
Post by Deleted on Jan 19, 2018 14:39:46 GMT -5
If you're going to play this card, you have to do it in amusing gif format please.
|
|