Fargo Breakdown - Singles @ BCAPL Nationals

[GideonF]

I agree. Statistics are a good thing and something that's been lacking in pool. Fargo does an okay job but I've found several holes which Mike choose to discredit and ignore. That's fine, it's his system. Waiting for someone to say the tired old pool line... "if you don't like it, make your own". Cause that helps...



There are many factors that are ignored, the biggest IMO is table size. One of thing is that playing speed isn't linear. 50 points for the lower ranked players is very different than 50 points for the higher ranked players. That's a huge flaw IMO. Hopefully Mike will take that into consideration and alter his system to becoming more accurate.

Mike hasn't ignored your views on table size. He has addressed them clearly. You may not accept his analysis, but you can't say he's ignored it.

As to the 50 points being different at lower levels than at the top level, my guess is that the answer is that the points aren't an absolute measure - it is a measure relative to players above and below them. Remember that the idea is that 100 Fargo point differential is supposed to equate to a 2:1 game differential.

 

[mikepage]

Mike hasn't ignored your views on table size. He has addressed them clearly. You may not accept his analysis, but you can't say he's ignored it.

As to the 50 points being different at lower levels than at the top level, my guess is that the answer is that the points aren't an absolute measure - it is a measure relative to players above and below them. Remember that the idea is that 100 Fargo point differential is supposed to equate to a 2:1 game differential.

Exactly. A 50-point gap at the low end is by definition "the same" as a 50-point gap at the high end. Here is EXACTLY what a 50-point gap means, wherever you find it.

Compute 2 to the power (50pointgap/100). That's 2^(50/100), which is 2^0.5, which is the square root of 2. So if you whip out your phone calculator and compute the square root of 2, you will see it is 1.41

1.41 is the ratio of games won by the upper player to games won by the lower player with a 50-point gap.

So that's 141 to 100, or 14 to 10, or 7 to 5

There are all kinds of things about the 50-point gap that might vary from the low end to the high end--what kinds of skill differences comprise it, how much dedicated practice it takes to close the gap, and so forth.

But what it means in terms of game-win propensity or chance of winning a particular set is the same...

 

[mikepage]

He was on the list for the platinum rated at 715, but is now not on the list and his Fargo is now in the 740's. Me and Alex olinger was chatting about this.


[...]

Lee --

Here is a post in this forum from around the time of the SBE. Shane Winters has been in the 740's for quite some time. I'm not sure what you are talking about, unless it was a clerical error or something.

 

[Treyf123]

Fargo rate will end up being a great thing IMO.. This year it is not IMO... My reasoning is I'm ranked a 640 which is my starter rank since my robustness is under 200... the only reason I'm there is I got moved from open to advance last year because my 8 ball team did very well yet I lost my first match in 9 ball and went 2 and out in the 8 ball singles... I've traveled a lot and played in the Q city tour and the GSBT on the east coast so just looking up some rankings from the better players that frequent those tours that get out and play in events that Fargo rate takes into consideration and I'm quite a bit over ranked... I was still able to put together a decent team and it's more of a vacation than anything so I'm still going... It will get better the more info gets put into it I hope.

Sent from my SM-G925V using Tapatalk

 

[mikepage]

Fargo rate will end up being a great thing IMO.. This year it is not IMO... My reasoning is I'm ranked a 640 which is my starter rank since my robustness is under 200... the only reason I'm there is I got moved from open to advance last year because my 8 ball team did very well yet I lost my first match in 9 ball and went 2 and out in the 8 ball singles... I've traveled a lot and played in the Q city tour and the GSBT on the east coast so just looking up some rankings from the better players that frequent those tours that get out and play in events that Fargo rate takes into consideration and I'm quite a bit over ranked... I was still able to put together a decent team and it's more of a vacation than anything so I'm still going... It will get better the more info gets put into it I hope.

Sent from my SM-G925V using Tapatalk

Trey, you are about three quarters of the way toward being established. About 6 matches from Tunica event in 2012, 5 matches from 2014 Vegas event, and 5 matches from 2015 Vegas event. 146 games total playing at 646 speed. Your starter rating of 625 has only a small influence, and that is what brings you down to 640.

Your record from your 5 matches last year in the Open singles is 22 wins, 13 losses

 

[Treyf123]

Trey, you are about three quarters of the way toward being established. About 6 matches from Tunica event in 2012, 5 matches from 2014 Vegas event, and 5 matches from 2015 Vegas event. 146 games total playing at 646 speed. Your starter rating of 625 has only a small influence, and that is what brings you down to 640.

Your record from your 5 matches last year in the Open singles is 22 wins, 13 losses

Thanks for your Quick response Mike I'm not sure how I had 5 matches unless both of the sets in 9 ball counted as a match. I do remember I won my first in losers bracket in the 8 ball so that would make 5 matches.. I do have one other question the Tunica event I played in was a handicapped event and I did very well but was getting 2 and 3 games on the wire from most of the competition just wondering how that's factored in that it was not an even race in most of those matches

Sent from my SM-G925V using Tapatalk

 

[Bob Jewett]

Thanks for your Quick response Mike I'm not sure how I had 5 matches unless both of the sets in 9 ball counted as a match. I do remember I won my first in losers bracket in the 8 ball so that would make 5 matches.. I do have one other question the Tunica event I played in was a handicapped event and I did very well but was getting 2 and 3 games on the wire from most of the competition just wondering how that's factored in that it was not an even race in most of those matches

...

In case Mike is busy...

If you are playing against a champion who is 100 rating points above you, and you beat him 3-6 in a handicapped match were you are only going to 3, neither of your ratings will move because that's the expected score for that rating difference. Your robustness would increase, of course.

If you beat him 3-4, you will go up because you got more games than expected against him.

If he beats you 7-2, you will go down because your loss was worse than expected.

One place where handicapped matches might go wrong for the ratings is if the games on the wire were included in the results as games won. They must not be included in the calculations -- only real games won are counted.

 

[yelvis111]

This is not a small piece; it is core to what we do.

I don't make claims about exact numbers. But within expected statistical fluctuations, yes.

85 points is 85 points

Below I show all matches we have between SVB and opponents around 85 points below him. I go up 7.5 points in both directions to get better statistics. There are 601 games, and Shane is supposed to win in a ratio of 1.80 to 1.

That ratio is Shane 386 games, opponents 215 games

The actual record is Shane 391, opponents 210

In other words, this group of players is performing at a level 85 points below Shane, as expected. And yes, we see this same kind of agreement when we look at 85 points for low-level league players. That is key to what we do.

The matches here are Shane's score, opponent's score

Mike Page,

Can you show Rob Saez's results against players 85 points below him?

Also, how long would it take for 2 Fargo ratings that were 100 points apart to converge, given identical results? In other words, if Player A started as a 425 and Player B started as a 525, and they played the next n games against the same opponents and posted the exact same results, what would n equal by the time they had the same rating? Would the starter ratings have any effect on the convergence point?

The Fargo ratings of their opponents should be immaterial to this question seeing as how their results indicate that they play at the exact same speed.

Taek Chang

 

[sbpoolleague]

Mike Page,

Can you show Rob Saez's results against players 85 points below him?

Also, how long would it take for 2 Fargo ratings that were 100 points apart to converge, given identical results? In other words, if Player A started as a 425 and Player B started as a 525, and they played the next n games against the same opponents and posted the exact same results, what would n equal by the time they had the same rating? Would the starter ratings have any effect on the convergence point?

The Fargo ratings of their opponents should be immaterial to this question seeing as how their results indicate that they play at the exact same speed.

Taek Chang

Don't you think Mike is busy?

And did it ever occur to you to say PLEASE?

 

[yelvis111]

Don't you think Mike is busy?

And did it ever occur to you to say PLEASE?

First of all, I don't think there's anything in my post/request that indicates that I'm pressing him for this data. This entire post has been a pretty free exchange of ideas and opinions about Fargo rate and its effect on the BCAPL Nationals. Mike has freely provided data in response to other requests earlier in this forum.

Second, who are you exactly? Are you Mike's nanny or knight protector?

{begin sarcasm font}If I were Mike Page, I'd feel warm and fuzzy knowing you were out there policing responses to my posts in OPEN forum.{end sarcasm font}

Taek Chang

 

[mikepage]

Mike Page,

Can you show Rob Saez's results against players 85 points below him?

Robb has only about one tenth the games in the system as SVB, so to do a similar analysis--which actually takes me considerable time--I'd have to take a much wider swath to get statistically meaningful results.

Also, how long would it take for 2 Fargo ratings that were 100 points apart to converge, given identical results? In other words, if Player A started as a 425 and Player B started as a 525, and they played the next n games against the same opponents and posted the exact same results, what would n equal by the time they had the same rating? Would the starter ratings have any effect on the convergence point?

The Fargo ratings of their opponents should be immaterial to this question seeing as how their results indicate that they play at the exact same speed.

Taek Chang

If by started at 525 and 425, you mean starter ratings of 525 and 425, then the number of games performing at the same level necessary to wipe out the difference is exactly 200. Beyond 200 games, that starter rating is completely ignored, and the rating is purely performance based.

 

[yelvis111]

Robb has only about one tenth the games in the system as SVB, so to do a similar analysis--which actually takes me considerable time

Beyond 200 games, that starter rating is completely ignored, and the rating is purely performance based.

No hurry on the Robb Saez data. I'm trying to get back to one of the original questions of the thread; namely, will the results from a 100 point gap between a 715 and a 615 mirror the results between a 615 and a 515?

As far as the second point, are you saying that the ratings will not converge until they get established at 200 games?

Once again for record, I'm a HUGE proponent of Fargo ratings.

Taek Chang

 

[mikepage]

No hurry on the Robb Saez data. I'm trying to get back to one of the original questions of the thread; namely, will the results from a 100 point gap between a 715 and a 615 mirror the results between a 615 and a 515?

This is something we've tested many times in many different ways. I will share some of that when I get the time.

As far as the second point, are you saying that the ratings will not converge until they get established at 200 games?

Yes. Suppose you and I actually both play the same and exactly like SVB --823 speed. You are given a starter rating of 525, and I am given a starter rating of 425. Then we both compete, playing like an 823. Below is exactly the trajectory our transitional ratings would take.

 

[alphadog]

Robb has only about one tenth the games in the system as SVB, so to do a similar analysis--which actually takes me considerable time--I'd have to take a much wider swath to get statistically meaningful results.




If by started at 525 and 425, you mean starter ratings of 525 and 425, then the number of games performing at the same level necessary to wipe out the difference is exactly 200. Beyond 200 games, that starter rating is completely ignored, and the rating is purely performance based.

THIS WHAT FORUMS ARE SUPPOSED TO BE ABOUT!:

Thanks to Mike Page for his great service to the billiard community!:wink::wink:

Thanks to yelvis111 for asking interesting,RELEVANT,questions!:wink:

 

[yelvis111]

This is something we've tested many times in many different ways. I will share some of that when I get the time.



Yes. Suppose you and I actually both play the same and exactly like SVB --823 speed. You are given a starter rating of 525, and I am given a starter rating of 425. Then we both compete, playing like an 823. Below is exactly the trajectory our transitional ratings would take.

Excellent.

Thank you Mike.

 

[sbpoolleague]

First of all, I don't think there's anything in my post/request that indicates that I'm pressing him for this data. This entire post has been a pretty free exchange of ideas and opinions about Fargo rate and its effect on the BCAPL Nationals. Mike has freely provided data in response to other requests earlier in this forum.

Second, who are you exactly? Are you Mike's nanny or knight protector?

{begin sarcasm font}If I were Mike Page, I'd feel warm and fuzzy knowing you were out there policing responses to my posts in OPEN forum.{end sarcasm font}

Taek Chang

I started this thread with a bunch of stats and analysis, and Mike has been gracious enough to add more stats and analysis. I am a league operator, that's all. Someone who is enthusiastic about Fargo Rate and the upcoming Nationals. Someone, like Mike, who spends a lot of time promoting Fargo rate as the next big thing to hit the pool universe.

I'm not Mike's nanny. Just someone who understands what is going on behind the scenes and is willing to help out in whatever way I can. I'm sorry if my post was a little gruff, but you are one of dozens of people who have asked Mike to provide custom stats and analysis FREE OF CHARGE and with nary a please or a thank you. Some of you have been gracious, but most feel they are entitled to such data.

Perhaps I need to dismount my high horse. If that's true then I will. I will be in Vegas this Thursday for nine days. If any of you see me in the casino or the tourney room, stop me and say hi. Would be great putting faces together with the AZB names.

 

[robsnotes4u]

Mike did mention to me he will also be there giving seminars on FargoRate. I am not sure what days


Sent from my iPhone using Tapatalk

 

[sbpoolleague]

BTW all singles players should compare the Fargo rating they see on the Fargo Rate Website with the one they see on the CSI Website.

In my league there are 8 players with well over 200 games of robustness yet the CSI website still shows them as unknown (525). This could mean a big change between Gold and Silver divisions, for example.

 

[T-dog]

BTW all singles players should compare the Fargo rating they see on the Fargo Rate Website with the one they see on the CSI Website.

In my league there are 8 players with well over 200 games of robustness yet the CSI website still shows them as unknown (525). This could mean a big change between Gold and Silver divisions, for example.

Do they have brackets drawn up yet and what software is used?

 

[sbpoolleague]

Do they have brackets drawn up yet and what software is used?

As of 1:45 pm Las Vegas Time no brackets are up. I would fully expect the 9-ball singles brackets up later today or early tomorrow, followed by the doubles brackets maybe on Wednesday.

CueSports Tournament System is the software used for the brackets. I have assumed that this software was written for CSI, but I have yet to ask Mark Griffin what the history of that software is.

----------------------------------------------------

After a little searching I found Mark's 2009 post about CTS....Ron Guilmette is the man behind CTS.

"Just so everyone knows, CSI (CueSports Int'l) has been developing a new software program for handling larger (and small) events. We first used this at the US Bar Table in Feb [2009]. It is known at CTS (CueSports Tournament System).

We currently only handle events that has our laptop on site - BUT after our May event, we wil get a new addition going. This will allow us to "loan" our software to any number of events.

The location will have to provide the names - the software can do the draw, table and time management, and anything else all the up to cutting the checks. The beauty is that any computer with internet access will be able to connect and have access to this software.

It is designed to handle many divisions simultaneously, and can handle unlimited number of tournaments at the same time.

I do not know the technical jargon on how this works - but this is unlike anything out there. I expect it to be functioning around Sept or so.

The plan is to make this available for free. The brackets will remain available for viewing for an unlimitged time.

I think this will be very cool. On site, it can be shown on a big screen or whatever.

If you have any questions, please contact either markg@playcsipool.com or ron@playcsipool.com."

Mark Griffin
CSI
BCAPL-USAPL-NCS