TaToH's strange match point phenomenon

[Umdeuter]

Match point in that post is clearly defined. It is the decider match of a series when both players are tied and the series has went to the last game.

no, it's all the games when he was able to close out the set. 2:0 up in a bo5, he had to lose 3 matchpoints. it is NOT decider matches.

rest of your post is correct anyway, but to be scientifically correct, you'd need to do much more calculations because the 16 (or rather 14) games are cherry-picked. you could arguably exclude the Jordan games and would be down to 12. then you may include more games from the past when he was for example able to beat Viper on a decider in Clown Cup. there also might be some contributing factors such as fatique or civ picking approach. still, it is a pretty crazy number which probably will hold to be significant under all realistic assumptions.

Well, if I understand correctly, the opponents he faced in those 16 match points are Viper (x4), Liereyy (x3), Hera (x3), Jordan (x2), Yo (x2), Saymyname (x1) and Villese (x1).

I understand your argument that Tatoh got to a match point against them in each case, but it still seems really difficult to give anyone a 50% chance on average of winning those 16 matches.

the spectacular thing here was that in the non-matchpoint games, he even won comfortably over 50% against the same opponents

In all of his 25 non-matchpoint-games, he won 15:10 games.

In his 14 matchpoint-games he lost 14 times. 0:14.

 

[Vossn]

So since there is a bit of confusion, and I'm not even certain myself anymore, I'll do a short recap of a) the sets he lost since HC4 (the date mentioned in the initial reddit post of Umdeuter), and b) the sets which went full length, which he won. Source which is used is https://aoe-elo.com/players
Considering I did all this by hand, there is a possibility I missed some sets.
a)
2-3 vs SayMyName; The Resurgence - Deciders
2-3 vs Villese; The Resurgence - Winners
0-2 vs Classicpro; OnlyBans - group stage, idk the tournament
0-2 vs Villese; Jordan's Medieval Brawl(JMB) season 1 #2 - Semi's
2-3 vs PROject_Belgium; Master of Socotra 2; K.O. Phase (I think ro16?)
4-5 vs Villese; Master of hyperrandom - Finals (Actually shown wrong on the site, luckily I remembered it went the full 9 games, and double checked liquipedia to be sure.)
2-3 vs TheViper; WWC - Semi's
3-4 vs Jordan; KOTD4 - Quarters
2-3 vs Liereyy; KOTD4 - Winners
2-3 vs TheViper; Holy Cup - Quarters
1-3 vs Liereyy; RBW5 - Quarters
1-2 vs DauT; RBW5 - group stage
1-2 vs Hera; RBW5 - group stage
1-2 vs Daniel; All-in cup - K.O. Phase
1-3 vs Slam; Masters of Arena 6 - Quarters
2-3 vs Yo; TOC - Semi's
0-3 vs Yo; Collection cup - Quarters
2-4 vs Vivi; RBW4 qualis
2-3 vs Hera; Arabia invitational 2 - Quarters
1-2 vs Bruh; Deep waters Gaming Pro League - Group stage
2-3 vs Hera; HC4 - Ro16

determination results a)
Out of 21 defeats, 15 have gone the full length of the set, as far as I can see, which is a large majority. I have not compared it to other top players though. Removing the bo3's, since those quickly rise in number with the likes of Deep waters Pro League, we get: 15 lost sets, 11 of which have gone the full length. This is pretty much the same ratio of 1.4 as when taking a look at the non bo3 sets.
b)
2-1 vs Dogao; Deep waters Pro League: Season 2 - Group Stage
2-1 vs ACCM; Deep waters Pro League: Season 2 - Group Stage
2-1 vs DauT; Deep waters Pro League: Season 2 - Group Stage
2-1 vs Jordan; Deep waters Pro League: Season 2 - Group Stage
2-1 vs DauT; RBW5 - Group Stage
2-1 vs Villese; RBW5 - Group Stage
3-2 vs MBL; TOC - ro16
3-2 vs MBL; Collection cup
2-1 vs PROject_Belgium; Empires Showdown 3
3-2 vs Miguel; RBW4 qualis
2-1 vs AngelinaJolie; RBW4 qualis
2-1 vs Nicov; Deep waters Gaming Pro League - Group stage
4-3 vs LaaaaaN; Rey de las américas 2
2-1 vs Capoch; Deep waters Gaming Pro League - Group stage
2-1 vs Dragonstar; Deep waters Gaming Pro League - Group stage
2-1 vs ACCM; Deep waters Gaming Pro League - Group stage
2-1 vs Miguel; Deep waters Gaming Pro League - Group stage
2-1 vs Vivi; Deep waters Gaming Pro League - Group stage

Determination results b)
In total there are 18 wins for TaToh in sets which have gone the full length. A lot of these however belong to the deep waters pro league, which is PA3. A lot of these are bo3's as well, which are very short sets, and I don't trust short sets. Removing them from the equation, the total amount of won sets which went full length becomes 4.

Conclusion:
Despite a slightly lower amount of sets going full length ending in a loss compared to a win (with a 11:4 ratio, for non bo3 sets), TaToh did manage to win a few sets which went to the last game. The last time he managed this though, was the Open Classic, which is over a year ago if I remember correctly. That's definitely a long time ago.

 

[TheShaunPlays]

Lmao midwits. Match point in that post is clearly defined. It is the decider match of a series when both players are tied and the series has went to the last game.

In my opinion it is absolutely clear that he chokes hard in these situations or something is up with his drafting that leaves him hopeless in the last game.

Let's put numbers to it.

I've read a couple of post above that he is 0-16.
We can model this as a Binomial process. Since we are looking only what is happening on those "match points" we can assume that the probabilty of him winning is 50% since to get there tatoh and his opponent have won the same number of games in the series. This approximation should be good enough for what we want to do here.

If he went 0-16... the probability of him winning more games than that i.e. 1-15, 2-14, 3-13, 4-12 etc. is an asthonishing 99,9984% so you decide if that is "bad luck" or a "random happening" or there is something else.

To put in other words if he replayed all those decider matches 100000 times in only two ocasions you would expect him going 0-16.

Match point in tennis is the point that wins the whole thing, not when both have a chance to win (I don't think it's possible for both players in tennis to have match point at the same time, I could be wrong though). Semantics though, we all know what the poster means

 

[TheShaunPlays]

If it were an issue of sports psychology, posts like this are probably not helpful.

I'm only a sports scientist but all the sport psychs I know would tell their athletes to not even look at posts online if there was a risk of negative impacts. It's a factor that is out of the athletes control should they decide to look, if they choose not to look then that is a form of control allowing for less variance in performance.

 

[Rayne]

I'm only a sports scientist but all the sport psychs I know would tell their athletes to not even look at posts online if there was a risk of negative impacts. It's a factor that is out of the athletes control should they decide to look, if they choose not to look then that is a form of control allowing for less variance in performance.

I thought I’d stay out of this topic, but from having a love for psychology, you’ve sparked my interest.

Isn’t this just avoidance though? And it never truly touches the issue and therefore no way to improve upon it? That’s given the information does show a reoccurring theme I’m sure Tatoh wants to get out of one way or another. Actually have my own thoughts on that but keeping them to myself.

 

[TheShaunPlays]

I thought I’d stay out of this topic, but from having a love for psychology, you’ve sparked my interest.

Isn’t this just avoidance though? And it never truly touches the issue and therefore no way to improve upon it? That’s given the information does show a reoccurring theme I’m sure Tatoh wants to get out of one way or another. Actually have my own thoughts on that but keeping them to myself.

I'm not a psych so I couldn't go into depth for you, 2 units of sports psychology does not an expert make. Generally though in sport we care about what we can control and don't care about what we can't control (this is not specific to psychology; reffing decisions, rep selections, paper talk, opposition etc).

 

[Potkeny]

Isn’t this just avoidance though? And it never truly touches the issue and therefore no way to improve upon it?

Maybe the reasoning is that the issue should be brought up by the expert when they deem the situation right for it, instead of reading it "randomly" in an online discussion?

That way the expert (and the athlete) can have better control of the impact of it, can schedule the work on it, and might be able to minimize the "damage" on results even if it didn't turn out how they wanted?

But I'm a complete amateur, just having fun thinking about it.

 

[Rayne]

Maybe the reasoning is that the issue should be brought up by the expert when they deem the situation right for it, instead of reading it "randomly" in an online discussion?

That way the expert (and the athlete) can have better control of the impact of it, can schedule the work on it, and might be able to minimize the "damage" on results even if it didn't turn out how they wanted?

But I'm a complete amateur, just having fun thinking about it.

Truthfully I don’t completely know about that.

I have never worked with an expert in psychology, and while I am not mocking the sports psychologists he mentioned, I just found the hypothetical advice of an avoidance strategy weird in any given scenario. That’s because issues never solve themselves, and could subtly hint at not wanting to confront it due to fear of it. But there may be something I am missing, and would have to talk to them about it.


Like one thing that concerns me, is that let’s say it is brought up in ‘the right scenario,’ changes don’t occur so quickly. It could take time, a lot of time in fact. And not just that, a psychologist can only go so far. People can only change something if they truly want to, so they have to seek that part themself. It’s their own journey that way and have to actively work on it for awhile.

So if this topic would bother someone having the exact issue, it may mean they are shying away from it. Psychology is all about breaking past falsity and staring at the truth about oneself with things, and that can be very difficult to do with a variety of things. If it were me at least, I wouldn’t mind looking at anything to find the answer to something I am trying to understand. Or maybe that’s just because I have found answers in the strangest of places, could just have a weird life like that.


Also while aoezone may have a bad track record of criticizing experts/known people harshly, this is probably one of the purer topics on these forums. Very unusual since I was expecting the opposite, but no one has mocked Tatoh or anything unless I missed it.

 

[TheShaunPlays]

Oh

Maybe the reasoning is that the issue should be brought up by the expert when they deem the situation right for it, instead of reading it "randomly" in an online discussion?

That way the expert (and the athlete) can have better control of the impact of it, can schedule the work on it, and might be able to minimize the "damage" on results even if it didn't turn out how they wanted?

But I'm a complete amateur, just having fun thinking about it.

Oh that's what he meant, yeah we would tackle the decider match issue head on if it was indeed a problem not a coincidence. But the comment I was replying to said "posts like this are not helpful", truth be told they shouldn't have an impact either way on Tati.

 

[TheShaunPlays]

Truthfully I don’t completely know about that.

I have never worked with an expert in psychology, and while I am not mocking the sports psychologists he mentioned, I just found the hypothetical advice of an avoidance strategy weird in any given scenario. That’s because issues never solve themselves, and could subtly hint at not wanting to confront it due to fear of it. But there may be something I am missing, and would have to talk to them about it.


Like one thing that concerns me, is that let’s say it is brought up in ‘the right scenario,’ changes don’t occur so quickly. It could take time, a lot of time in fact. And not just that, a psychologist can only go so far. People can only change something if they truly want to, so they have to seek that part themself. It’s their own journey that way and have to actively work on it for awhile.

So if this topic would bother someone having the exact issue, it may mean they are shying away from it. Psychology is all about breaking past falsity and staring at the truth about oneself with things, and that can be very difficult to do with a variety of things. If it were me at least, I wouldn’t mind looking at anything to find the answer to something I am trying to understand. Or maybe that’s just because I have found answers in the strangest of places, could just have a weird life like that.


Also while aoezone may have a bad track record of criticizing experts/known people harshly, this is probably one of the purer topics on these forums. Very unusual since I was expecting the opposite, but no one has mocked Tatoh or anything unless I missed it.

I misunderstood what you meant, we wouldn't avoid the decider match issue if it was indeed an issue. What we would avoid is public fan forums if the player has a temperament that means reading stuff like this could negatively impact them. Plenty of athletes don't use social media for this very reason, just reading it could lead to them believing it for some people. Others can read it with no impact, and others use it as motivation, it's very individualised.

 

[Rayne]

I misunderstood what you meant, we wouldn't avoid the decider match issue if it was indeed an issue. What we would avoid is public fan forums if the player has a temperament that means reading stuff like this could negatively impact them. Plenty of athletes don't use social media for this very reason, just reading it could lead to them believing it for some people. Others can read it with no impact, and others use it as motivation, it's very individualised.

Ah, with you all there. I misunderstood too. If someone does have that mindset it would indeed be an issue. I stay away from most social media for that reason, so many things that can throw anyone into a hissy fit if they’re looking for one tbh 11.

 

[Potkeny]

the comment I was replying to said "posts like this are not helpful"

Now I have no idea if I replied the right way (and/or if I misunderstood you/Rayne), but seeing as we all more or less agree with each other (or at least think we understand what the others tried to say), I guess it doesn't really matter anymore 11

 

[_IamBronco]

I will give it a shot too...

So as I understand the question we are trying to find out if Tatoh performs better or worse than usual in matchpoint games. I collected the data from all the 1v1 S-tier tournaments he has played in (from Escape Gaming Euro Cup until the group stage of The Resurgence). In total these are more than 300 games and 75 sets for Tatoh. For the biggest part these sets are against other top players.

To have a reference I also extracted the data from the following players: TheViper, Hera, Liereyy, JorDan, Yo, Nicov, DauT and MbL. I choose these players without a reason other than these were the first that came into my mind. The reason is, that we want to compare their performance with Tatoh's.
(Note that Tatoh did not play in every tournament but in most of them)

Disclaimer: The sample is relatively small, but at least it gives a first overview wether anomalies occur.

Here are the results:

We should look at the relation between games won and mathpoints won. This gives us the following diagram. The x and y-axes are in percentages. (x-axis: Games won, y-axis: matchpoint games won)

First, we see here that I chose the sample of players poorly. We should take Jordan out because he skews the sample. This is because Jordan has had many breaks in recent years.

(x-axis: Games won, y-axis: matchpoint games won)

Without Jordan we get:

Despite the small sample size, a fairly clear trend line emerges. Tatoh has a 0.54 win rate in tournament games and a 0.56 win rate in matchpoint games.

If Tatoh were playing worse than normal in matchpoint games, we should see a clearer deviation here. But the fact is that Tatohs matchpoint performance perfectly lines up with his usual performance in other tournament games.

For me it was far more interesting to look at this diagram:

Here we see how many matchpoint games are played on average per tournament set. This is basically an indicator of which games we should pay special attention to in future tournaments.

Daut's sets are by far the most exiting, since he donates his opponents on average at least one matchpoint game and has one mathpoint himself, whereas Vipers sets are the most boring to watch...

The Lord is the one who thinks of the common plebs and gives us the entertainment we deserve.

 

[Umdeuter]

seems like a strong indicator for it being a random thing.

can you show how that developed over time for tatoh? and maybe how it looks vs top 3 players vs other players? (basically: where was the overperformance of him when we know the underperformance that we saw before)

 

[TopperHarley]

I will give it a shot too...

So as I understand the question we are trying to find out if Tatoh performs better or worse than usual in matchpoint games. I collected the data from all the 1v1 S-tier tournaments he has played in (from Escape Gaming Euro Cup until the group stage of The Resurgence). In total these are more than 300 games and 75 sets for Tatoh. For the biggest part these sets are against other top players.

To have a reference I also extracted the data from the following players: TheViper, Hera, Liereyy, JorDan, Yo, Nicov, DauT and MbL. I choose these players without a reason other than these were the first that came into my mind. The reason is, that we want to compare their performance with Tatoh's.
(Note that Tatoh did not play in every tournament but in most of them)

Disclaimer: The sample is relatively small, but at least it gives a first overview wether anomalies occur.

Here are the results:

We should look at the relation between games won and mathpoints won. This gives us the following diagram. The x and y-axes are in percentages.
View attachment 200691
First, we see here that I chose the sample of players poorly. We should take Jordan out because he skews the sample. This is because Jordan has had many breaks in recent years.

Without Jordan we get:
View attachment 200692
Despite the small sample size, a fairly clear trend line emerges. Tatoh has a 0.54 win rate in tournament games and a 0.56 win rate in matchpoint games.

If Tatoh were playing worse than normal in matchpoint games, we should see a clearer deviation here. But the fact is that Tatohs matchpoint performance perfectly lines up with his usual performance in other tournament games.

For me it was far more interesting to look at this diagram:
View attachment 200693
Here we see how many matchpoint games are played on average per tournament set. This is basically an indicator of which games we should pay special attention to in future tournaments.

Daut's sets are by far the most exiting, since he donates his opponents on average at least one matchpoint game and has one mathpoint himself, whereas Vipers sets are the most boring to watch...

The Lord is the one the one who thinks of the common plebs and gives us the entertainment we deserve.

Thanks for this! Doesn’t this show that Tatoh is actually pretty good at match point games?

Nicov, Daut and MBL all have weaker win rates in matchpoints than in general, while Tatoh seems to almost excel in matchpoints (his winrate there is even better than his winrate in general).

 

[_IamBronco]

Thanks for this! Doesn’t this show that Tatoh is actually pretty good at match point games?

Nicov, Daut and MBL all have weaker win rates in matchpoints than in general, while Tatoh seems to almost excel in matchpoints (his winrate there is even better than his winrate in general).

Sorry, that was maybe a bit missleading.

The diagram shows the overall win rate on the x-axis and the matchpoint win rate on the y-axis. (i will edit it in the post.)

So actually You could say that MbL overperforms a bit. But again the database is so small that it is all within the margin of error. My point was just, that it looks like Tatoh is perfectly fine in the trend that is basically just what you would expect: the players who win more games also win more matchpoint games.

 

[TopperHarley]

Sorry, that was maybe a bit missleading.

The diagram shows the overall win rate on the x-axis and the matchpoint win rate on the y-axis. (i will edit it in the post.)

So actually You could say that MbL overperforms a bit. But again the database is so small that it is all within the margin of error. My point was just, that it looks like Tatoh is perfectly fine in the trend that is basically just what you would expect: the players who win more games also win more matchpoint games.

I don't think it is misleading, i think its spot-on!

According to my interpretation, MbL is rather UNDERperfoming in matchpoints isnt he? (with a winrate of ~0.6 in general but with only ~0.52 in matchpoints)

But i agree in general, that the data is too small to really draw conclusions from and i agree that tatoh has an expected winrate in matchpoints (and nothing that would make you think he struggles in these situations more than anyone else).

 

[_IamBronco]

According to my interpretation, MbL is rather UNDERperfoming in matchpoints isnt he? (with a winrate of ~0.6 in general but with only ~0.52 in matchpoints)

It is the other way around. The headline line in the diagram was messed up and i changed it.

When a player is above the trend line then he is "overperforming" (with respect to the small database)

Here is the summary and additionally the "matchpoints denied" (so the won games where the opponent had a matchpoint)

Here we see that DauT definitely a bit above the trend and MbL kinda underperforming compared to other players.

 

[_IamBronco]

seems like a strong indicator for it being a random thing.

can you show how that developed over time for tatoh? and maybe how it looks vs top 3 players vs other players? (basically: where was the overperformance of him when we know the underperformance that we saw before)

Here are the diagrams over time for individual players. I picked DauT and MbL for comparison because they where quite close to Tatoh in the previous diagrams. But I had to exclude 2018 from Dauts data, since there was only NAC and his graphs would basically be just at zero (https://liquipedia.net/ageofempires/Nilis_Apartment_Cup/1)
MbL has no S-Tier tournament games in 1v1 in 2022, so there are only 5 years for him as well.

So You could conclude that 2021 was one of Tatoh weaker years in terms of "matchpoint performance", but lets see how it developes over 2022. For me there are just too few games to make a statement about it.

 

[Vossn]

An anomaly or not, it keeps happening for some reason...

 

[IYIyTh]

This theory also assumes all players improve equally and that aoe2 does not have an exceptionally high skill cap and gap at this level of play. Lierrey and Viper are not going to surrender advantages the same way Nicov or Tatoh might in a consistent manner not for a lack of trying but the fact you're comparing different levels of skill, even at smaller and smaller ranking disparity. For example, no one has issue with Jordan losing to Hera in Game 7 when most woukd agree Hera is the better player, but Tatoh and Nicov (slightly but distinctly further down that pecking order,) is a cause for analysis?


Also, I'm not sure Nicov/Tatoh match is a good example, since someone had to win.

 

[LowEloNobody]

This theory also assumes all players improve equally and that aoe2 does not have an exceptionally high skill cap and gap at this level of play. Lierrey and Viper are not going to surrender advantages the same way Nicov or Tatoh might in a consistent manner not for a lack of trying but the fact you're comparing different levels of skill, even at smaller and smaller ranking disparity. For example, no one has issue with Jordan losing to Hera in Game 7 when most woukd agree Hera is the better player, but Tatoh and Nicov (slightly but distinctly further down that pecking order,) is a cause for analysis?


Also, I'm not sure Nicov/Tatoh match is a good example, since someone had to win.

Well, Jordan and Nicov don't have a long-running history of making it to the very last match of a series and losing. They have quite a few more 4-2, 3-1 style losses.

 

[IYIyTh]

Well, Jordan and Nicov don't have a long-running history of making it to the very last match of a series and losing. They have quite a few more 4-2, 3-1 style losses.

But aren't they consistently ranked lower than the opponents they're playing as well?

There is nothing wrong with being a solid 7/8, but these sports are played over long series, so upsets are less likely (and should be so, arguably more than other sports.)

 

[LowEloNobody]

But aren't they consistently ranked lower than the opponents they're playing as well?

There is nothing wrong with being a solid 7/8, but these sports are played over long series, so upsets are less likely (and should be so, arguably more than other sports.)

Sure, but I think the point is that there are no other major players who have the same track record with match points

 

[Liamsmalley7]

I forgot about this thread! I think a lot of the analysis above actually suggests he isn't as bad during deciding matches as it seemed at first glance. Although, I couldn't help but be reminded of it during Only Land cup