HUSNG Statistics Project
 

I have the results of over 900k HUSNGs and 132k players that span several months of this year. Rather than sit on this information I'd like to contribute to the forum by compiling useful statistics on... everything.

The things I do have:
- Buyin
- Time of Day
- Result from 2 and 4 man games

Things I do not have:
- A lot of data from the tubos and non-increasing blinds
- Durations

I need 2+2's help to determine what data to extract and how to do it. For example, you want to compare $50s and $100s? How would you do it? You want to know what winrates are achievable? How would you do it? Winning percentage quartiles or what? Want to know what time of day the most fish play? Come up with the rules to do it and I'll get the results.

All ideas are welcome, even if you cant think of how to do it.

Nichomacheo

Re: HUSNG Statistics Project
 

Nicho, is it something you gathered yourself, or are you using sharkscope or something like that? Also, is this coming from all buy-ins, including the highest?

Re: HUSNG Statistics Project
 

Gathered by my own means, not from SharkScope.
Everything from $5s to $5000s.

Re: HUSNG Statistics Project
 

Calculate how much money pokerstars has made.

~rob

Re: HUSNG Statistics Project
 

[ QUOTE ]
Calculate how much money pokerstars has made.

~rob

[/ QUOTE ]

Haha

Re: HUSNG Statistics Project
 

You gathered the data from 900k HUSNGs by yourself?

Very impressive. I cant figure out how to do that. Must be some kind of sharkscope manager. [img]/images/graemlins/laugh.gif[/img]

I use the Tourneymanager (and PT of course) so the time with the most fish is from 18pm to 3am (ET time). My sample size for Hu is quite small.

With a big sample size from a lot of players I would add the stats from the best players and then calculate an average winrate for the different buyins.

slimbob

Re: HUSNG Statistics Project
 

I've got some neat ideas for u if u have time to chat. As u know, this is what i do for a living. I'll try to post something tomorrow. This is a good database that you have.

Long story short, you need significant adjustments before comparing rates across pools of players.

Indy(PS if u want me to analyze the data in SAS, you can send it to me in excel or something)

Re: HUSNG Statistics Project
 

Indy~

I'm nicho's roommate.... OBV... anyways, there's a lot of [censored] data.

Last I knew it took >2 hours to back up the database. (Which he does every other day... weird)

Slimbob... he said no sharkscope. I don't know HOW he finds the time while still playing a dozen games or so a day to record thousands of games. Such tedious work... and he still passes his classes. AMAZING!

~rob

Re: HUSNG Statistics Project
 

Looking forward to this. Don't have any ideas beyond the obvious queries different posters have mentioned though. Good work! Goldmund

Re: HUSNG Statistics Project
 

I think the best way to analyse these is by stakes, then by player, then by winrate.

I would do this as follows, with these objective in mind:

<font color="blue"> Things we want to discover:

Whether the higher levels are in fact harder than the level below, or whether the people playing them just have more money. Ie - Are there particular levels which winning players should be targeting because the conditions are particularly favourable (maybe the 50s or 100s)?

Or whether certain lower levels are harder than their counterparts higher up, and should be skipped if possible.

For example, there is a school of thought that NL cash ring games are harder at the $50 level than at the $100 level because NL50 players have a higher % of graduates from the NL25 level, whereas richer fish just dive straight in at the NL100 level. Therefore, you should be taking shots at NL100 is your bankroll can possiblity tolerate it, and playing as little as possible at NL50. Similar lessons for HU would be very useful.</font>

Step 1

Search for all the players in the database playing at a certain level.

Step 2

Split this list into two separate lists: one for those who have played 100 or more games at that level (for ease of reference, the "pros"), one for those who have played less (the "amateurs").

<font color="green"> Expected results: as the buy-in rises, the number of "pros" should rise, and the number of "amateurs" should fall. </font>

<font color="red"> Conclusion 1: At x level, x % of players are amateurs, and x % are pros. The higher the % of amateurs, the easier that level should be. </font>


Step 3

For each player on each list, work out their winrate (ie % of games player that they won). Separate them into bands - I would suggest &lt;70%, 70-65, 65-60, 60-55, 55-52.5, 52.5-50, 50-45, 45-40, &gt;40. Tabulate how many players fall into each band.

<font color="green"> Expected results:

1) as the buy-in rises, there should be more players in the middle bands and less in the outer bands as the average skill of the player pool rises.

2) the amateurs should have a higher concentration in the lower bands, the pros in the higher bands (ie the amateurs should be the fish, and the pros the sharks)</font>

<font color="red"> Conclusion 2: ratio of fish to sharks at each level. Also, when compared to the results from stage 2, how many of the so-called "pros" have reached their 'level of incompetence' (ie have moved up to stakes that they cannot easily beat, and so are "pro fish" - our favourite sort of player [img]/images/graemlins/cool.gif[/img]!) </font>

Step 4

This may be a bit complex, but here we go. Look for players that appear on lists at more than one level, then ask of each player:

1. What is the highest level they have played at?
2. What is the lowest level they hae played at?
3. What level have they played the most games at?
4. What is the highest level they have played more than 100 games at?
5. Have they beaten (ie had a winrate of 60% or more) all the lower levels they have played at over more than 100 games?

<font color="green">Expected Results: I really don't know. But we are looking to see how many players are building their way up by learning their trade at the lower levels before moving up, and how many players randomly pick a buy-in depending on their mood and see how they get on. Also, we may find out what levels get too high for people who are just playing about, and start only to attract "serious", winning players.</font>

There's probably a lot more conclusions we could draw from the data in this step, but I can't think of it yet.

Let me know what you think of these ideas!

Re: HUSNG Statistics Project
 

[ QUOTE ]

I have the results of over 900k HUSNGs and 132k players that span several months of this year. Rather than sit on this information I'd like to contribute to the forum by compiling useful statistics on... everything.

The things I do have:
- Buyin
- Time of Day
- Result from 2 and 4 man games

Things I do not have:
- A lot of data from the tubos and non-increasing blinds
- Durations

I need 2+2's help to determine what data to extract and how to do it. For example, you want to compare $50s and $100s? How would you do it? You want to know what winrates are achievable? How would you do it? Winning percentage quartiles or what? Want to know what time of day the most fish play? Come up with the rules to do it and I'll get the results.

All ideas are welcome, even if you cant think of how to do it.

Nichomacheo

[/ QUOTE ]

Another thing you might try to do is rank the players (or even compute Elo-type ratings as in chess) rather than just look at a simple statistic like 1) ROI or 2) dollars/day. The problem with 1 is that it will favour people who avoid the strong players, while 2 will favour the grinders. This may only be possible with HU games.

Working out who is the 'strongest' SnG player and seeing where you rank sounds like much more fun.

(I'm trying to do this for heads-up cash games.)

Marv

Re: HUSNG Statistics Project
 

Ok here are my thoughts. Not having duration is huge because the big ? with HU matches is whether or not a fast and aggro style with a lower ROI is better $EV than a cautious style with a higher ROI that takes longer per match. Because you have so few variables, we will have to view this as a tabular/observational task. My interest here would be to evaluate the "learning curve" effect. Bascially, within each limit of play I would look at the overall win rate, then look at the win rate after removing the first 100 games played per player, then look at it after removing the first 200 games played, etc etc...Naturally, this analysis should be on players with at least 400 or so games played.

Another analysis we should do is a prediction of winrate in a logistic model. I'd set it up to regress win rates against buyin(mandatory), time of day(perhaps categorized into morning, evening, night), and number of games played by this player. From this we would get very useful info like the amount of % win/loss we can expect by playing under different conditions.


Indy

Re: HUSNG Statistics Project
 

[ QUOTE ]
Must be some kind of sharkscope manager.

[/ QUOTE ]

Not Sharkscope at all. This is not what I want to talk about, so lets stick to the project.

[ QUOTE ]
then calculate an average winrate for the different buyins.

[/ QUOTE ]

This is a popular request, but, eh, its 50% across the entire population. Think about it: for every 1 win, there is 1 loss. Four-man games will be different though, but thats not what you asked.

[ QUOTE ]
PS if u want me to analyze the data in SAS

[/ QUOTE ]

Dont know what SAS is, but, the data is not in a standard Access database. I had to create my own directory hasing scheme to make it fast.

[ QUOTE ]
I don't know HOW he finds the time while still playing a dozen games or so a day to record thousands of games

[/ QUOTE ]

Its called multitasking, jackass. [img]/images/graemlins/wink.gif[/img]

[ QUOTE ]
I think the best way to analyse these is by stakes, then by player, then by winrate.

[/ QUOTE ]

Penguin, great response as always. I was thinking about doing something like this and I'm glad you've put some thought into it. I'm going to spend my operations research class today thinking about it and with everyone's help I want to do something like what you were saying. At a high level, it would be breaking the winrates up for each buyin (ie what % of players at the $50s have a winrate 45-50%, 50-55% etc). Just what Penguin said.

[ QUOTE ]
Another thing you might try to do is rank the players (or even compute Elo-type ratings as in chess) rather than just look at a simple statistic like 1) ROI or 2) dollars/day.

[/ QUOTE ]

I dont know what ELO is, but I understand what you're saying. You want to come up with a number that represents skill level. I've tried long and hard to do this, but its amazing complex. If you can come up with a smart way to do it that'll factor in the different buyins etc, then I'd do it.

[ QUOTE ]
My interest here would be to evaluate the "learning curve" effect. Bascially, within each limit of play I would look at the overall win rate, then look at the win rate after removing the first 100 games played per player, then look at it after removing the first 200 games played, etc etc...Naturally, this analysis should be on players with at least 400 or so games played.

[/ QUOTE ]

I dont think this would be informational. My data is not 100% complete for any period of time. Its a sampling of 900k games out of... who knows. I probably have 60-75% of the game results. Any analysis that involves profit would be pointless (ROI wouldnt). I dont think games 0-100 of a sample would be any different from 200-300 other than variance. Unless you follow a player's ENTIRE results (without blanks), you couldnt do the analysis you're talking about.

[ QUOTE ]
Another analysis we should do is a prediction of winrate in a logistic model. I'd set it up to regress win rates against buyin(mandatory), time of day(perhaps categorized into morning, evening, night), and number of games played by this player. From this we would get very useful info like the amount of % win/loss we can expect by playing under different conditions.

[/ QUOTE ]

I like this, but you'd have to explain how to do it.

Re: HUSNG Statistics Project
 

Not sure if any of the previous posters have suggested this, but tracking the win% of sets of players with 100+ game samples at different levels seems interesting. That way you get to see how a 63% win-player on the 50$ buy-in level performs on the 100$ level etc. Goldmund

Re: HUSNG Statistics Project
 

[ QUOTE ]
Not sure if any of the previous posters have suggested this, but tracking the win% of sets of players with 100+ game samples at different levels seems interesting. That way you get to see how a 63% win-player on the 50$ buy-in level performs on the 100$ level etc. Goldmund

[/ QUOTE ]

I could do that. I'm not sure how many people have 100+ games at multiple levels. I'm sure there are plenty, but how many do I need to have a good result?

Re: HUSNG Statistics Project
 

I'm thinking about something like this:

For each buy-in level, create this graph, in which each unique player is represented by a point on on a 2D graph: X axis for how many games he played, and Y axis for his % win rate (0-100). Now on a Z axis we can see how many players have same certain X AND Y (i.e, in cases when more than one point is in the same place on the X,Y graph).

It is a 3D graph basically. I hope this is doable and that it makes sense.

Re: HUSNG Statistics Project
 

[ QUOTE ]
I'm thinking about something like this:

For each buy-in level, create this graph, in which each unique player is represented by a point on on a 2D graph: X axis for how many games he played, and Y axis for his % win rate (0-100). Now on a Z axis we can see how many players have same certain X AND Y (i.e, in cases when more than one point is in the same place on the X,Y graph).

It is a 3D graph basically. I hope this is doable and that it makes sense.

[/ QUOTE ]

I understand what you're saying, but I think its overkill. You cant look at something like this and easily understand the results. Two-dimensional graphs are much more user-friendly and what I think we should shoot for.

Re: HUSNG Statistics Project
 

[ QUOTE ]

I understand what you're saying, but I think its overkill. You cant look at something like this and easily understand the results. Two-dimensional graphs are much more user-friendly and what I think we should shoot for.

[/ QUOTE ]

Perharps this is true, but I believe that the big majority of interesting information will be found on the x,y graph layer, also considering the fact that for players who played more than few dozens of games you will rarely have exact same X AND Y. Most of the the information on the Z axis will therefore "exist" only in the area where you'll find players who played a very small amount of games, which is the least interesting area anyway. Think how improbable it is to have 2 players who played _exactly_ n games (say n&gt;100), and who at the same time have _exactly_ the same winrate.

Re: HUSNG Statistics Project
 

How's this sound to everybody?

For every player in the database, check to see which of the buyins he has more than 100 games for. For each buyin, determine the winrate and increment a variable that corresponds to that buyin/winrate. Simply:

$50s,45%-50% = $45%-50% + 1

At the end, I'll have data on every buyin and what winrates are most likely and how they are skewed.

Hows this sound?

Re: HUSNG Statistics Project
 

[ QUOTE ]
what winrates are most likely and how they are skewed.

[/ QUOTE ]

I think that this is the most interesting data for every buy in. The question should be how to weight different "sample sizes" coming from different players.

Re: HUSNG Statistics Project
 

will you also list which players have what winrates at what levels, etc?

Re: HUSNG Statistics Project
 

[ QUOTE ]
I think that this is the most interesting data for every buy in. The question should be how to weight different "sample sizes" coming from different players.

[/ QUOTE ]

I dont know. Anyone?

[ QUOTE ]
will you also list which players have what winrates at what levels, etc?

[/ QUOTE ]

No.

Re: HUSNG Statistics Project
 

[ QUOTE ]
For every player in the database, check to see which of the buyins he has more than 100 games for. For each buyin, determine the winrate and increment a variable that corresponds to that buyin/winrate.

[/ QUOTE ]

Of how many players do you have 100+ gamesamples at different levels? 900K games will not generate a great number of these I think. Goldmund

Re: HUSNG Statistics Project
 

[ QUOTE ]
[ QUOTE ]
For every player in the database, check to see which of the buyins he has more than 100 games for. For each buyin, determine the winrate and increment a variable that corresponds to that buyin/winrate.

[/ QUOTE ]

Of how many players do you have 100+ gamesamples at different levels? 900K games will not generate a great number of these I think. Goldmund

[/ QUOTE ]

I have no idea. A few hundred, maybe.

Re: HUSNG Statistics Project
 

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
For every player in the database, check to see which of the buyins he has more than 100 games for. For each buyin, determine the winrate and increment a variable that corresponds to that buyin/winrate.

[/ QUOTE ]

Of how many players do you have 100+ gamesamples at different levels? 900K games will not generate a great number of these I think. Goldmund

[/ QUOTE ]

I have no idea. A few hundred, maybe.

[/ QUOTE ]

What % of the players in your database are regulars, and what % of games played have a regular playing?

oh and how much better are regulars, and what times/days of the week have more regulars?

Re: HUSNG Statistics Project
 

Stars does not allow data mining, but if somehow all this information is from stars, I may be interested in purchasing it. Either way, thanks for doing it and I look forward to the results

Re: HUSNG Statistics Project
 

[ QUOTE ]
Stars does not allow data mining

[/ QUOTE ]

They dont allow you to datamine hands. I datamined results for HUSNGs, which, like SharkScope, is allowed at the moment.

PM me.

Nichomacheo

Re: HUSNG Statistics Project
 

New suggestion: a comparison of winrates of a set of players in the Regs v. the Turbo's. To clarify: compare the winrate of winning players ABC etc in the reg's with that in te Turbo's. This might shed some light on the regs/turbo's-issue. Goldmund

Re: HUSNG Statistics Project
 

Like I said, I dont have much data from the Turbos because they are relatively new and I havent datamined the results in over 6 weeks.

Re: HUSNG Statistics Project
 

[ QUOTE ]
Want to know what time of day the most fish play?

[/ QUOTE ]

Yes please?

Re: HUSNG Statistics Project
 

How about how much tilt affects a player. For each specific player, you could determine the odds of a win after a win, after one loss, 2 losses, etc. and see if a player's win percentage actually does go down after previous losses (or goes up after a winning game).

Re: HUSNG Statistics Project
 

hey man, any updates on your progress or anything?

i love u nicho,
acidca

Re: HUSNG Statistics Project
 

this seems like an amazing endeavor, im ver inerested to see what you coe up with. im sure we are all very thankful of this

Re: HUSNG Statistics Project
 

Soon ... been busy with other things.

Re: HUSNG Statistics Project
 

bump. nicho what happened to this?

Re: HUSNG Statistics Project
 

Other projects.