Updated FargoRates are out

mikepage

AzB Silver Member
Silver Member
I'm not sure I see 1 player on that list ranked under her
That she is the favorite over .

The answer to me--as it so often is--is more data, and I don't see a lot of that coming soon. Chezka (just off the bottom of that top 100 list) came through the US last fall, and Siming played in the Austria Open last month. So this here is actually all we have for these two playing 750+ males.

Siming went 14 to 14 against opponents rated an average of 774
Chezka went 25 to 32 against opponents rated a weighted AVERAGE OF 780

It's only a few games, but Siming is performing in the 770s and Chezka is performing in the 750s

It would be good to have several hundred more games like this.
 

Attachments

  • simchez.png
    simchez.png
    12.4 KB · Views: 338

The-Professor

AzB Silver Member
Silver Member
How important is the Cue Ball?

I find it hard to believe that Siming Chen would be favored to beat Donnie Mills in a long set of rotation period!
 

BeiberLvr

AzB Silver Member
Silver Member
The answer to me--as it so often is--is more data, and I don't see a lot of that coming soon. Chezka (just off the bottom of that top 100 list) came through the US last fall, and Siming played in the Austria Open last month. So this here is actually all we have for these two playing 750+ males.

Siming went 14 to 14 against opponents rated an average of 774
Chezka went 25 to 32 against opponents rated a weighted AVERAGE OF 780

It's only a few games, but Siming is performing in the 770s and Chezka is performing in the 750s

It would be good to have several hundred more games like this.

Wait, Chezka beat Donny?

But Siming plays better than Chezka.

And Donny wants to play Siming?
 

donny mills

AzB Silver Member
Silver Member
Wait, Chezka beat Donny?

But Siming plays better than Chezka.

And Donny wants to play Siming?

Lol this makes me cringe. A race to 11 in a tournament doesnt mean much. Stan Shuffett beat me 11-2 in the us open in 2011, tommy kennedy just knocked me out of the us open 11-7 almost 2 weeks ago. Will I spot stan or tommy the last 3 in a long set? Yes I will.
 

one stroke

AzB Silver Member
Silver Member
You are in a club with a lot members ;-)

Yes that club is large , tournament play is 1/4 mile like cars players are set up for that we see this yr in a yr out with the MC team , not one player beats Shane in a mile race period,, Donnie is a mile specialist so throw 1/4 mile analytics in the trash there worthless in projecting long races ,


1
 

Oze147

AzB Silver Member
Silver Member
But the fact that there is no single dominant player in a short race format is what suggests the format is a crapshoot.

When you have a 100+ players in an event and claim the cream always rises to the top, your 'cream' is 1 of a group of superior players.

I would not want to bet against someone else being able to pick 1 of the top 3 players for the us open, allowing him a group of 20 players.

Same bet with group of 1? Action, jackshun.

Or maybe it is not crapshoot but sport?

Tennis had three very dominant players over the last 20 years. Federer, Nadal and Dokovic. These three are the only players who won multiple US-Open titels. In the last 20 years there were 13 different US-Open winners.

In Golf the US-Open was won by 17 different players in the past 20 years.

So following your logic, we should play best of 11 tennismatches and every golf course should have 54 holes.

And even if we would change the formats in pool to 50+ races you wouldn't get the one and only player who shines above all the others.
You would get a group of 20-30 players who are really good at playing day-long pool matches. Just like we have it today.
 

Poolplaya9

Tellin' it like it is...
Silver Member
Here's some nonsense you posted from Post #119

"No, what is fairest is a neutral racker who racks with a triangle rack. That way you both get the same racks,"

Do explain how they would get the same racks using a triangle rack? Have you even played pool before?

Have you ever used good reading and comprehension before? You didn’t this time. I didn’t say every single rack would be absolutely perfect (not that they necessarily are with the magic rack either, and not that it couldn't be possible with the triangle and good balls, good table, and good racker). What I said is that both players would be getting the same racks as each other with a neutral racker. Even if both players were making their best effort, the racks from two different people are never going to be as similar as the racks from just one person. That’s not nonsense, that’s a fact. And if one of the people had a different racking goal than the other (which always has a reasonably high likelihood of happening in a big gambling match), like say one was attempting to manipulate the rack while other was not, then they would be even less similar yet. Any way you cut it racks from only one person are going to be more similar to each other than racks from two different people and this holds true with any rack although some will show more difference than others.

On a side note I could have also suggested the neutral racker use the magic rack, which would have also given both sides the same racks as each other, and I don’t have a huge problem with that, but I suggested the triangle rack because I think it gives a much better chance for pool skills to determine the winner here, particularly in 9 ball and particularly when one of them is known for their prowess in reading/breaking the magic rack. It’s just too easy for somebody to figure out a great break for the magic rack, make the same ball in the same pocket over and over, often leaving very similar and easy patterns over and over, and that just wouldn’t tell you much of anything about who the better player is, and that was the whole point, to make a statement about who the better player is.
 

Poolplaya9

Tellin' it like it is...
Silver Member
It was months ago. I have no idea what thread. The thread had something to do with / morphed into a legal discussion. I had someone who was arguing with you on that thread pm me and said they thought you sounded like a cop, and I agreed. After that I actually indicated it in that thread I believe. I seem to remember an affirming response, but for sure no denial. In another thread a couple of weeks later you mentioned " ticketing someone ". That was what I was mostly basing it on. Most recently; a week or two ago, in a response to you I called you " a so called cop " with again no denial. Bottom line it does not matter to me one little bit if you are or are not. I actually have no problems with cops, I have many cop buddies - some very close buddies. The whole reason it was brought up because the discussion was something along the lines of the moralities of the pool culture and my opinion as well as many others were the only people that felt the way you were presenting yourself were leos.

This is a perfect example of why you get so many things so wrong, and why there are so many things that you are unable to understand. It’s because you don’t choose to read carefully and try to comprehend what you read, and because you are always forming beliefs without any regard for evidence or good logic, and because you jump to conclusions without thinking things out. Obviously people have differing cognitive abilities but a lot of it is a choice and it seems that you don’t even make an effort to try to think things out or comprehend what you read. It’s mental laziness. Let me set you straight point by point.


It was months ago. I have no idea what thread. The thread had something to do with / morphed into a legal discussion.
Didn't happen. I have never said or insinuated that I was a cop.

I had someone who was arguing with you on that thread pm me and said they thought you sounded like a cop, and I agreed.
Given the evidence you and he jumped to a silly conclusion. That being said, I've had someone PM me and tell me they thought you sounded like you were a gay porn star, and I agreed based on things you have said. Instead of telling everybody on here several times that this is your occupation as if it were a fact because I and others believe it though, I chose to keep it to myself instead because A) it doesn't have anything to do with pool and wasn't necessary to the discussions and this is a pool forum not a women's gossip forum, B) the only reason I would have for mentioning it since it isn't pool related or necessary to the discussion would be as an attempt at a personal attack, and C) I didn't think that it "seeming" "feeling" and "sounding" true to me and others was good enough evidence. By the way, it does not matter to me one little bit if you are or are not. Get some of the points yet?

After that I actually indicated it in that thread I believe. I seem to remember an affirming response, but for sure no denial.
You don't even remember if you mentioned it in that thread or not, but somehow you are sure about what I affirmed or denied in that thread even though you don't even know if you mentioned it in that thread? Not sure you thought this one out either...

For the record, I have denied it at least 3 times that you have mentioned it. Unlike what comes from your imagination, I can actually point to those posts. Again, not that I find the accusation to be an insult, because being a cop is more compliment than insult in my book, but I hate when people use bad logic or make conclusions based on crappy evidence and often call people on those things which is why I chose to call you on it when you jumped to that firm conclusion and spewed that nonsense with zero to support it.

In another thread a couple of weeks later you mentioned " ticketing someone ". That was what I was mostly basing it on.
That didn't happen either. Like as in me writing traffic tickets? Seriously, I'm not sure where you get this stuff from but there is zero chance that any sane and reasonable person could get that firm belief from anything I have ever said. This is on the same level as my former legit schizophrenic male neighbor who among other things swore to me that burglars were breaking into his house several times a week for the sole purpose of moving his underwear to different locations around the house and to "eat a bowl of his soup" (but they never ever took a single thing even though there were some nice valuables in plain sight including on one occasion a decent sum of cash on the coffee table right where the imaginary burglar had sat to eat a bowl of his soup...lol). Wait, are you my old neighbor? I might know you! Do you remember the guy that would break in several times a week to move your clothes around and eat your soup but always left all your valuables in place?

Most recently; a week or two ago, in a response to you I called you " a so called cop " with again no denial.
Pretty sure I denied it that time too. There is a small chance this is one I didn't see or just chose not to respond to but it is more likely you missed my response or at least you imagined that you did since you seem to do a lot of imagining.

Bottom line it does not matter to me one little bit if you are or are not.
If you didn't think it mattered you wouldn't feel the need to bring it up over and over now would you? Plus you clearly said it with a negative connotation and intending it to be an insult particularly initially so again there goes your "it doesn't matter to me" argument.

The whole reason it was brought up because the discussion was something along the lines of the moralities of the pool culture and my opinion as well as many others were the only people that felt the way you were presenting yourself were leos.
So cops are the only ones that have certain morals? Like I said, you really have to learn to use your head and think things out a little bit...
 
Last edited:

chefjeff

If not now...
Silver Member
If races went to 10,000 no rating system would work.

In fact, pool wouldn't exist at all. Who the hell wants to play to 10,000?

People, normal people--you know with jobs and a life--- play short matches and take their wins and loses from that.

Might as well play one game of 1-Pocket as play a race to 10,000. (sorry, couldn't help myself)



Jeff Livingston
 

cardiac kid

Super Senior Member
Silver Member
After fifteen pages of this unanswerable question, the only thing that matters is who the real pool world recognizes as the best. Of course that would be Jeanette Lee. Black Widow. Lady in black. What ever the players who count think her name is. She hasn't won a tournament in years yet she is instantly recognizable to the "player on the street". What else really matters? Carry on. Move along, nothing here to see.

Lyn
 

poolscholar

AzB Silver Member
Silver Member
Mike, are you able to provide a rating ranges for a given confidence %, for Chen and Mills?

You frequently mention of the possibility that ratings are +- 20 points off their listed rating, so it would be nice to see what the actual values are. I assume there can be a wide range especially for people without a lot of data.
 

donuteric

always a newbie
Silver Member
You frequently mention of the possibility that ratings are +- 20 points off their listed rating, so it would be nice to see what the actual values are. I assume there can be a wide range especially for people without a lot of data.

My impression is that the FargoRate's robustness is meant as a proxy to give some indication of confidence, although currently it's simply displaying the number of games a player has in the system. It'd be useful if robustness is scaled between 1-100 to suggest how confident an assigned rating is.
 

JohnnyOzone

AzB Silver Member
Silver Member
On a side note I could have also suggested the neutral racker use the magic rack, which would have also given both sides the same racks as each other, and I don’t have a huge problem with that, but I suggested the triangle rack because I think it gives a much better chance for pool skills to determine the winner here, particularly in 9 ball and particularly when one of them is known for their prowess in reading/breaking the magic rack. It’s just too easy for somebody to figure out a great break for the magic rack, make the same ball in the same pocket over and over, often leaving very similar and easy patterns over and over, and that just wouldn’t tell you much of anything about who the better player is, and that was the whole point, to make a statement about who the better player is.

I agree with everything you are saying, but I would submit to you therefore that the pros should be playing a different game where the break shot is not such a huge part of the result of the match, if the true aim is to determine who is the better player ( since 9 ball, and to a lesser extent 10 ball, mostly only proves who is the better breaker)
 

mikepage

AzB Silver Member
Silver Member
You frequently mention of the possibility that ratings are +- 20 points off their listed rating, so it would be nice to see what the actual values are. I assume there can be a wide range especially for people without a lot of data.

The uncertainty in the ratings is a complicated problem. It's not like the uncertainty in the outcome after a certain number of coin flips or bernoulli trials--though we do expect some relation to that. A single game against a similarly rated opponent carries more information than does a single game against a much stronger or much weaker opponent. A game against an opponent who has 600 games in the system carries more information than a game against an opponent who has 300 games or 30 games. A game against an unrated opponent tells nothing about you.

(Brief gobblygook in this paragraph and then it gets understandable again next paragraph). What we maximize is the log-likelihood function, and that gives us the MLE (Maximum Likelihood Estimates) for the ratings. The Hessian matrix--the second derivative of log-likelihood--evaluated at the optimum ratings is what is called the observed Fisher Information Matrix. There is a theorem (Cramer Rao Bound) that relates the diagonal elements of the inverse of the Hessian to the variances/covariances--that's where we are going, analyzing that. But we're not there yet.

In the meantime we can get some brute force estimates of the standard errors under actual conditions. Here is an example. We took a player who has played a very large number of games (5500) in the last two years and has a rating of 622. Then we can take a random sampling of, say, 200 of those games and compute a rating just based on the 200 games. Then we do this many times. What we find is the standard error (standard deviation) of the 200-game rating is about 20 points.

That suggests when a rating is based on 200 games, just at the threshold of what we refer to as "established," it has about a two-thirds chance of being within 20 points of the "true" rating and about a 95% chance of being within 40 points.

These intervals decrease as you log more games with the square root of the number of games. So if you want to cut the uncertainty in half (68% chance you are within 10 points and 95% chance you are within 20 points) you have to go to 800 games.

The average number of games for players on the top 100 list is 3000. So in general when you look at those numbers it is fair to think of them as probably (68%) right within 5 points. And someone fairly new--like Aranas--is probably right within 10 points.
 
Last edited:

donuteric

always a newbie
Silver Member
The uncertainty in the ratings is a complicated problem. It's not like the uncertainty in the outcome after a certain number of coin flips or bernoulli trials--though we do expect some relation to that. A single game against a similarly rated opponent carries more information than does a single game against a much stronger or much weaker opponent. A game against an opponent who has 600 games in the system carries more information than a game against an opponent who has 300 games or 30 games. A game against an unrated opponent tells nothing about you.

(Brief gobblygook in this paragraph and then it gets understandable again next paragraph). What we maximize is the log-likelihood function, and that gives us the MLE (Maximum Likelihood Estimates) for the ratings. The Hessian matrix--the second derivative of log-likelihood--evaluated at the optimum ratings is what is called the observed Fisher Information Matrix. There is a theorem (Cramer Rao Bound) that relates the diagonal elements of the inverse of the Hessian to the variances/covariances--that's where we are going, analyzing that. But we're not there yet.

In the meantime we can get some brute force estimates of the standard errors under actual conditions. Here is an example. We took a player who has played a very large number of games (5500) in the last two years and has a rating of 622. Then we can take a random sampling of, say, 200 of those games and compute a rating just based on the 200 games. Then we do this many times. What we find is the standard error (standard deviation) of the 200-game rating is about 20 points.

That suggests when a rating is based on 200 games, just at the threshold of what we refer to as "established," it has about a two-thirds chance of being within 20 points of the "true" rating and about a 95% chance of being within 40 points.

These intervals decrease as you log more games with the square root of the number of games. So if you want to cut the uncertainty in half (68% chance you are within 10 points and 95% chance you are within 20 points) you have to go to 800 games.

The average number of games for players on the top 100 list is 3000. So in general when you look at those numbers it is fair to think of them as probably (68%) right within 5 points. And someone fairly new--like Aranas--is probably right within 10 points.

Thank you for the excellent explanation. Out of curiosity, if you find an estimator that does not attain the min variance of CRLB, what do you do?
 
Top