So with what you stated, Shane is a 91% favorite over John playing to 150 points in straight pool?
Fargo Rating? Valley vs Diamond
- Thread starter Belford
- Start date
So with what you stated, Shane is a 91% favorite over John playing to 150 points in straight pool?
When I first moved back to OKC and started playing regularly I was lost on the Diamonds they have there with the fast cloth and bouncy rails. I completely understand your feeling.
Last year in league we went to some places with valley tables - I was lost again - pockets were way easier but the way the cueball moved was "off".
I think that Fargo takes all this into account by not adjusting for any of it. In other words how you play on different equipment IS part of your overall rating because your performance across equipment is averaged in a sense because the data only looks at games won and lost. The toughness of the tournament in terms of equipment and depth of field don't mean anything.
So in the purest sense Fargo ratings tell you who you are more accurately. Yes I might play much better on one type of equipment and worse on another but that then is the reality of my ability.
On the surface it does seem that if a player is ranked a 600 who exclusively plays on 9ft tables plays a player who is ranked 600 who plays exclusively on 7ft tables that the 9ft player would be "better" on that table.
Mike Page said that's not the case though when that scenario has been tested against the data.
I am reminded of a story in Buddy Hall's book, Rags to Rifleman. (I can check it now that I have a copy again)
In it a game is set up between Buddy Hall and Jim Rempe for 25k on a 9ft table. Buddy is the barbox champion and Rempe is the 9ft champion. It's 10 ahead for 25k.
Buddy breaks first and runs 6 racks and then breaks dry. Rempe breaks and runs out and then breaks dry. Buddy runs out and runs four more to win the set. Table time .45cts.
Last edited:
O e other thing I forgot to mention..i think playing ln different types and size tables will make you an all around better player
I think I am fortunate compared to a lot of league players.since I play in several apa divisions I get to play on al types of tables
Great shape valleys
Shitty valleys
Gold crowns
7'diamonds
My best 8 ball game ever was just 2 weeks ago on a diamond. Won 4-0.vs another 5 in 5 innings. Had one break and run . 2 innings ended I. Safes by me . And 2 innings were a result of me scratching on the break twice.
My best 9 ball game was recently where I beat a s/l 9 on the gold crowns.
Like I say..if you are able to adapt the type tables you pls on will not affect your rating
I think I am fortunate compared to a lot of league players.since I play in several apa divisions I get to play on al types of tables
Great shape valleys
Shitty valleys
Gold crowns
7'diamonds
My best 8 ball game ever was just 2 weeks ago on a diamond. Won 4-0.vs another 5 in 5 innings. Had one break and run . 2 innings ended I. Safes by me . And 2 innings were a result of me scratching on the break twice.
My best 9 ball game was recently where I beat a s/l 9 on the gold crowns.
Like I say..if you are able to adapt the type tables you pls on will not affect your rating
FargoRate counts games won not balls pocketed. I count money in my pocket not balls pocketed. I recall one player being on tilt so bad, he ran to the 7 every rack and left me 3 ball outs over and over. In a spot shot contest he owned me. In a stroke shot contest he owned me.
And just because I could not "Out Shoot" them does not mean I could not shoot.
lol... I think I have played John Barton 3 times
One was at the Cue Club in vegas--pretty normal
another time was on his triple-shimmed valley bar box--super tight pockets--crazy.
And it went downhill from there. The third time was on I think a 7' diamond that had cloth that had colorful psychedelic, glow-in-the-dark designs all over it...
so don't you be complaining about fast cloth and bouncy rails ;-)
John and I are within about 10 points of one another--but we knew that already...
That's not a lot of variation. Do you think it might be different next season?
With something that is roughly a 50-50 proposition like flipping a coin and you take, say, 100 tries, you expect the result to have some random variation. In fact if you got exactly 50 heads every time you flipped a coin 100 times, we could make a lot of money in Vegas.
So, how much variation is expected with a coin flip? For 100 flips it will be more lop-sided than 45-55 something like 32% of the time. It will be more lop-sided than 60-40 about 5% of the time.
While your friend's results do suggest that he plays better on a smaller table, they are also within the window of being simple random variations from identical true percentages.
COUPLING!!!!
The different table sizes have already been accounted for by the coupling of players that play on both. I'm sure Mike has the data to prove this point much like he did with the awesome example of the two Karen Corr's.
As far as John Schmidt is concerned, I had no idea is rating was so low (739), relatively speaking. I see why his data poses a bit of a conundrum.
The different table sizes have already been accounted for by the coupling of players that play on both. I'm sure Mike has the data to prove this point much like he did with the awesome example of the two Karen Corr's.
As far as John Schmidt is concerned, I had no idea is rating was so low (739), relatively speaking. I see why his data poses a bit of a conundrum.
The real question is,
What is the chance of Bob Jewett beating Shane in a game of straight pool?
If you accept the very shaky hypotheses that a rack of straight pool is equivalent to a rack of nine ball and that Fargo Ratings are equivalent in the two games, then yes.
On the other hand, you would also conclude that a player rated at 562 for nine ball would have 0% chance of beating Shane to 125 at straight pool, but such things happen.:grin:
JC
Coos Cues
I have noticed some inconsistencies lately with the ratings. I watched a streamed match last weekend and looked up the players and both were 40-50 points higher than my 606 with sufficient robustness. I wanted to gamble with both of them ASAP.
I think there are regional fluctuations regardless of how we are told it's supposed to work.
JC
I think there are regional fluctuations regardless of how we are told it's supposed to work.
JC
Andrew I was just about to say the same thing to you...that's the major flaw with your argument. You don't have to agree or even understand how FargoRate works but that doesn't change facts.
That is not a number of games. That is 177 matches played over the course of several years. Say the average match consists of 6 games played because they are handicapped races. That is over 1,000 games played. He is more than likely winning quite a few more games on the 8 foot table vs 9 foot table.
If I get really bored, then I will see what the games won vs lost for him on each size table actually is. I used him as an example because he has a large sample size of matches on different size tables.
Somebody always has a chance! I think a 0% chance would be the result of a forfeit.
Andrew, how can you shun what could be the best thing that happened to pool in decades simply because you are too stubborn to admit that you don't understand how it works? There are some really complicated mathematics at work here. I have a Masters degree in applied mathematics and it is hard for me to follow.
This has been explained before, including in a two-column series I did for Billiards Digest. Briefly:
Match scores are gathered. The system has over a million games for over 30,000 players recorded. The number of games for each player in each match is recorded.
The relative ratings for a pair of players is calculated according to all the matches between them. If A has a win ratio of 2:1 against B, A is rated 100 points above B. If B then has a 2:1 ratio against C, B is 100 points above C and A is 100 points above C.
Now, if we also have some matches between A and C, things get a little more complicated. Based on the info above A would be expected to beat C by a 4:1 ratio. Instead suppose it is only 3.5:1. Then the ratings are adjusted to give the most likely set of ratings that would produce the results seen. The number of pair-wise games is taken into account, so if A and C have played only 9 games (7-2 result) and the A-B and B-C pairs have hundreds of games, the A-C result will only move the ratings a little.
That's pretty simple for three players. To calculate the ratings for 10s of thousands of players is slightly more complicated. That's what the cluster of computers is for. Every day they calculate the most likely ratings for all of the players considering all of the matches.
And here is the FargoRate FAQ http://www.fargorate.com/#faq
Last edited:
I know what we need to explain things to you. An animated Venn diagram. I'll get on that and get back to you soon.
The best thing to happen to pool in decades is a pretty laughable statement. I have no doubt Mike has put a lot of work into this and if all things were equal, it would probably be pretty amazing and accurate. What Im saying is, not all things are equal.
As I've stated, Mike has not explained how the system works, so of course I do not know exactly how it works. Just the same, Mike doesn't know what my favorite pair of underwear is because I haven't told him. Both are the best kept secrets in pool and arguably the best things to happen to pool in decades.
But the expected variation result still holds for the 177 match events. It doesn't make any difference that each has several games within it. If you do run the numbers, I think you will find that the percentages of game wins are a lot closer (after adjusting for handicaps) than the percentages of match wins.
I think the idea that he hasn't explained it is pretty laughable. The math is not short and simple (like for some league handicaps) to make it transparent enough to be useful for us to see it. But the logic is pretty clear: Games won/lost, 100 pts = 2-1, recalculated for everyone daily. Information about Elo systems in general is widely available.
(I do think you're right about players whose ratings are based on 7-footers being overestimated relative to players whose ratings are based on 9-footers.)
Last edited: