Fargo Rating?

tucson9ball

tucson9ball

AzB Silver Member
Silver Member
About a month ago somebody posted on Facebook what my rating was, but for the life of me I can't find it. I went to the Fargo site and there is no info under my name.
How do we get our Fargo rating?
Does somebody have a link to get ratings?
 
tucson9ball said:
Thanks, I have a rating of 679 with a robustness of 207?

What is robustness?
Click to expand...

The robustness is the sample size. Obviously the more matches you have in the system, the more accurate your rating will be. I can't recall what Mike Page is using for his robustness number, whether it be the number of matches or individual games or whatever, but it is on the main Fargorate.com site if you look around.
 
tucson9ball said:
Thanks, I have a rating of 679 with a robustness of 207?

What is robustness?
Click to expand...
I think it's the number of games. They consider robustness trustworthy over 500 games as a sample size, but 207 is a fair sample size I'd think.

The 679 suggests you'd get beaten about 2-1, or say 10-5 on average by most top 10 US pros, other than SVB, who are around the 780 mark.

That's how I understand it anyway. 100 points difference equates to 2-1 score differential.

Colin
 
Phishing attack?

I clicked on the link and my anti-malware software popped up a "Phishing Attack" warning.

Be careful.
 
I just want to say how much I appreciate the work that has gone into developing the Fargo system. I can only imagine how difficult the data mining is for a project like this.

One question that comes to mind: How would Fargo rates be used when comparing a male vs female player? In your videos you use Jasmin and Earl as an example. It seems to me like the data sets used to develop their scores would be mostly isolated to either sex. Sure, there are players like Karren Corr who happily play in the men's events, but the sexes are still significantly isolated from each other. If the ratings are based on how you play against peers, and your peers are mostly of the same sex, how can it be that female rates and male rates are comparable?

I understand that for a player like Karren this isn't a problem. She has enough data playing against the men that her score is valid when compared against another male player. However, I can't help but feeling like a lot of female players do not fall into this category.

I guess my real question is this: Are there enough "Karren" players out there to bridge the gap between the two sexes, from a statistical perspective? Please do not hesitate to correct any false assumptions. Thank you!
 
Prince H said:
I just want to say how much I appreciate the work that has gone into developing the Fargo system. I can only imagine how difficult the data mining is for a project like this.

One question that comes to mind: How would Fargo rates be used when comparing a male vs female player? In your videos you use Jasmin and Earl as an example. It seems to me like the data sets used to develop their scores would be mostly isolated to either sex. Sure, there are players like Karren Corr who happily play in the men's events, but the sexes are still significantly isolated from each other. If the ratings are based on how you play against peers, and your peers are mostly of the same sex, how can it be that female rates and male rates are comparable?

I understand that for a player like Karren this isn't a problem. She has enough data playing against the men that her score is valid when compared against another male player. However, I can't help but feeling like a lot of female players do not fall into this category.

I guess my real question is this: Are there enough "Karren" players out there to bridge the gap between the two sexes, from a statistical perspective? Please do not hesitate to correct any false assumptions. Thank you!
Click to expand...

You are correct. That is indeed an issue, and you should not consider male and female ratings perfectly comparable.
 
Disagree

railbird99 said:
You are correct. That is indeed an issue, and you should not consider male and female ratings perfectly comparable.
Click to expand...

Disagree. If the male and female results are sufficiently connected gender has nothing to do with it.
 
SWRLE said:
Disagree. If the male and female results are sufficiently connected gender has nothing to do with it.
Click to expand...

They are not sufficiently connected, that's the point. They are loosely and indirectly connected at best. I didn't say they were completely disconnected, but it's enough to distort the rankings between genders.
 
railbird99 said:
They are not sufficiently connected, that's the point. They are loosely and indirectly connected at best. I didn't say they were completely disconnected, but it's enough to distort the rankings between genders.
Click to expand...

This can become extremely accurate over time. There are many Top level female players that play in Open events such as Jasmine, Karen, Monica, and Vivian. There are also mediocre level players that participate in Open events albeit less often.

There are numerous women of all skill levels that participate in local weekly tournaments all over the nation. If Fargo gathers more data at the local level then this will be sufficient enough because men who have established ratings will be playing women who have established ratings.
 
Not only does Karen (and Jennifer and Jeanette, etc.) play men, but then she also plays other women. I believe there are enough connections to make them comparable - but there don't need to be a ton, the system will still find the optimal solution. It's the same issue for local groups of players who don't get out much. A few will, and if those few are well connected within their group, that's all you need to get a very accurate score.
 
Rhea said:
This can become extremely accurate over time. There are many Top level female players that play in Open events such as Jasmine, Karen, Monica, and Vivian. There are also mediocre level players that participate in Open events albeit less often.

There are numerous women of all skill levels that participate in local weekly tournaments all over the nation. If Fargo gathers more data at the local level then this will be sufficient enough because men who have established ratings will be playing women who have established ratings.
Click to expand...

I'm not saying it will remain an issue, but it is currently an issue.
 
BRussell said:
Not only does Karen (and Jennifer and Jeanette, etc.) play men, but then she also plays other women. I believe there are enough connections to make them comparable - but there don't need to be a ton, the system will still find the optimal solution. It's the same issue for local groups of players who don't get out much. A few will, and if those few are well connected within their group, that's all you need to get a very accurate score.
Click to expand...

The ELO system is completely relative. It will be inaccurate when you take a woman that is completely isolated from playing with men and compare her ranking with men rankings. Yes that woman may have played against another woman who has some scores against men. This helps in making the scores more comparable, but it is still far from perfect.

"but there don't need to be a ton, the system will still find the optimal solution"

The "system" is not going get a magical solution out of thin air. The inputs to the formula rely on wins and losses between players. When not many of these inputs exist that connect the two isolated groups, you aren't going to get one cohesive set of comparable scores. It's that simple.
 
Last edited:
Prince H said:
One question that comes to mind: How would Fargo rates be used when comparing a male vs female player? ... It seems to me like the data sets used to develop their scores would be mostly isolated to either sex. Sure, there are players like Karren Corr who happily play in the men's events, but the sexes are still significantly isolated from each other. If the ratings are based on how you play against peers, and your peers are mostly of the same sex, how can it be that female rates and male rates are comparable?
Click to expand...

There was some discussion of this topic in the following thread within the last few days.
http://forums.azbilliards.com/showthread.php?t=407252
Mike Page (who built Fargorate) gives his opinion on the topic and about and how he feels the system is currently doing in regards to linking the men with the women in post #110 of that thread. The way I read it his opinion is essentially that the male and female players may not be absolutely perfectly linked yet and will continue to get better over time but that the system has already done a pretty good job linking them and that the women's fargo ratings are already accurate enough to be able to make fairly accurate comparisons with male players.
 
The fargo system is in its infancy. We can learn from the history of chess since it's elo system has been around far longer.

After a couple minutes of googling, I found a source of this exact problem in chess. It's very similar to the situation in pool. Most women play almost exclusively in women events, while some play against men. They actually manually added 100 points to all female players because of this issue:

http://www.3-dbaseball.net/2015/05/gender-in-chess-part-2-elo-ratings.html

5th paragraph
 
You must log in or register to reply here.
Back
Top