Johnny Archer and Fargo Ratings

[jsp]

Mike, I just introduced myself to your ratings system by watching your first "overview" youtube video. I think the concept for the Fargo rating system is simply ingenious. Awesome job. I hope it takes off even more than it already has.

Although older games are valued less as a normal part of the optimization...

Without completely revealing your secret sauce, can you provide a more quantitative explanation as to how much older games are valued less in your system?

I'm wondering if you ever thought about another category of your rating system in which it only rates the latest 200 games (arbitrary number) and completely ignores any games prior. This new category would be roughly equivalent to a season's batting average in baseball, compared to one's career batting average. Or maybe your app can compute the ratings based on the last x games? (Maybe your app already does this?)

My concern is the ratings for someone how has been playing well below his potential for years but finally solved something in his game and became a world beater in the last few months. The Fargo rating system may be very slow to account for his dramatic change in speed.

 

[thebark]

What is the Fargo rating of Billy Pilgrim?

 

[mikepage]

Mike, I just introduced myself to your ratings system by watching your first "overview" youtube video. I think the concept for the Fargo rating system is simply ingenious. Awesome job. I hope it takes off even more than it already has.

Thanks Joe

Without completely revealing your secret sauce, can you provide a more quantitative explanation as to how much older games are valued less in your system?

Sure, there is an exponential decay in time with a half-life of 3 years. So a 3-year-old game is worth half of one today, and a 6-year-old game is worth one fourth.

I'm wondering if you ever thought about another category of your rating system in which it only rates the latest 200 games (arbitrary number) and completely ignores any games prior. This new category would be roughly equivalent to a season's batting average in baseball, compared to one's career batting average. Or maybe your app can compute the ratings based on the last x games? (Maybe your app already does this?)

We have in our development stage tested many things like this. If you and I played 20 games two years ago, we can't throw them out for you without throwing them out for me as well. They might be unimportant for you but important for me. There are a number of trade-offs. We have thought a fair amount about this

My concern is the ratings for someone how has been playing well below his potential for years but finally solved something in his game and became a world beater in the last few months. The Fargo rating system may be very slow to account for his dramatic change in speed.

It is not. In determining our balance between statistical significance and timely data, we paid careful attention to top young players, for whom we have games at 14-years old, 15, 16, 17... These players change faster than most.

Bottom line is we have a pretty good balance between the quality of our prediction of player improvement and the quality of our prediction of player performance.

While most of the discussion now has been about high-level players, our system is designed for ALL of us.

 

[BasementDweller]

Wow.

I'm so impressed.

This is just what pool needed and it may very well play a large part in the resurgence of the game we all love.

I love it. Great work.

 

[one stroke]

Wow.

I'm so impressed.

This is just what pool needed and it may very well play a large part in the resurgence of the game we all love.

I love it. Great work.

Really hows that ?


1

 

[BasementDweller]

Really hows that ?


1

It could help bridge the gap between not only ALL of the different and competing amateur leagues but also the gap between the pros and the amateurs.

 

[BRussell]

It's interesting to compare Archer and Darren Appleton - they have almost identical ratings (788 & 789), but Appleton has won every big tournament in sight in the last 5 years. Is Archer just unlucky? Does Appleton know how to win the big ones?

 

[AtLarge]

Mike -- You always refer to "we" when talking about work on the ratings system. Can you tell us who else you are working with on this?

 

[Lynch]

By looking at Archers matches/scores, it looks like the bulk of the matches he lost, especially against the big guns were only by a couple games. I know he got smoked a couple times, but it seems to me that this would account for his high ranking. He wins a lot against the mid tier and top pros and when he loses it's generally close. If he would lose with more lopsided scores against the top guys, his rating would be much lower. This is what I'm seeing in regards to the system, if I'm reading it right.

I really like what this system has to offer and look forward to it's future. It must be a pain to gather all this information, but the hard work should be worth it in the end. Mike has a great passion for the game and is doing something that is really positive for the game IMO.

 

[jsp]

Thanks Joe



Sure, there is an exponential decay in time with a half-life of 3 years. So a 3-year-old game is worth half of one today, and a 6-year-old game is worth one fourth.



We have in our development stage tested many things like this. If you and I played 20 games two years ago, we can't throw them out for you without throwing them out for me as well. They might be unimportant for you but important for me. There are a number of trade-offs. We have thought a fair amount about this



It is not. In determining our balance between statistical significance and timely data, we paid careful attention to top young players, for whom we have games at 14-years old, 15, 16, 17... These players change faster than most.

Bottom line is we have a pretty good balance between the quality of our prediction of player improvement and the quality of our prediction of player performance.

While most of the discussion now has been about high-level players, our system is designed for ALL of us.

Mike, thanks for the reply. I'm sure questions that pop in my head have already thought about by you in much greater detail.

Has your algorithm already be set in stone, or are you continuously tweaking it here and there? Is your algorithm open source somewhere so others can see exactly how things are calculated?

Mike -- You always refer to "we" when talking about work on the ratings system. Can you tell us who else you are working with on this?

At the bottom of http://www.fargorate.com/ there is a tab in blue labeled "Who We Are". Clicking on it shows that Steve Ernst is also involved.

 

[robsnotes4u]

I would like to thank Mike. It is very impressive when someone answers every question quickly and professionally. Knowing him, and discussing Fargo Rate with him, Mike, puts a lot is thought into everything until he goes forward. Look at how many years he has worked on this until it has been brought to everyone. Just look at the website to see the enormous amounts of data.

It has been awesome playing under Mike's ratings for the last 4 plus years. Being able to watch your rating to see improvement is perfect for me.

Thanks Mike, once again you have done something to promote the game.


Sent from my iPhone using Tapatalk

 

[9BallKY]

At last year's Bigfoot challenge, as one example, Archer lost first round 11-8 against SVB. Probably nobody remembers this. But the fact is no other player got past 8 against Shane for the rest of the tournament. If Archer just happened to have a different draw and got to the finals of that event, we would all be talking about him very differently. When by performance it's not very different.

Actually the one shot where Johnny hooked himself could have made a difference in that set.I don't remember the score but it was like a one game difference in the score at that time. Shane won that game and if I remember correctly he broke and ran the next rack which is a 3 game swing.

 

[spartan]

So Batman does beat Superman...

It is blasphemous that that Synder guy even think of matching Batman v Superman
Batty Bat doesn't stand a chance
Everyone knows that Supes can fly and at speed of light, can reverse time

Oh sorry we are talking pool here

 

[Kris_b1104]

It is blasphemous that that Synder guy even think of matching Batman v Superman
Batty Bat doesn't stand a chance
Everyone knows that Supes can fly and at speed of light, can reverse time

Oh sorry we are talking pool here

Guess you didn't know the movie's based off the comic book storyline of The Dark Knight Returns by Frank Miller. Superman gets his fair share of beating

 

[rivercitysledge]

Where is brokeman archer? He made less than 7k this year....thats my list

 

[mikepage]

I have a question about your system.


Can one match drastically change a layers rating?


Like for example his win over Lee Van Corteza. It says that he beat Lee Van 11-1... if he would have beat Lee Van 11-10 would that have changed his overall rating much?

I just took out the 11-1 match with Corteza altogether. Johnny's rating goes down by 0.9 points and Corteza's rating goes up by 1.4 points as a result.

For players with fewer games in the system the impact would be larger.

 

[Kid Dynomite]

It is designed to find an individual's performance rating and to accurately predict how a single matches outcome will be.

To get it to do what you and I think several others are wanting it to be, there would have to be an algorithm included weighting placement in tourneys as well as games and matches won against specific opponents.

The danger in that though is that draw plays a much bigger role in the distance gone in tourneys and it becomes nigh on impossible to create an accurate algorithm without insane amounts of manual input and opens it up to inputer discretion more.

I think this is a very accurate system and as more people adopt it, it will become more and more accurate.

Jaden

It is SMOKE and MIRRORS!!!

Nothing has been provided regarding HOW the numbers are reached???

Is the US OPEN worth the same as the MEMPHIS OPEN???

One event is worth 20K and another worth 4k. The person with the most is the best!!!

Archer is FAR from the top of the leader board for money and/or points. But, we are suppose to take someones word for it that Archer is the number 2 player. Matchroom and Mosconi Cup and BCA leadership have all been lobbying for some ranking system to slip in on the billiard community and repeatedly there are folks trying to GAME the system. This post is just primer for the fargo system to be adopted. Its like putting lipstick on a pig. IMO

KD

 

[iusedtoberich]

It is SMOKE and MIRRORS!!!

Nothing has been provided regarding HOW the numbers are reached???

Is the US OPEN worth the same as the MEMPHIS OPEN???

One event is worth 20K and another worth 4k. The person with the most is the best!!!

Archer is FAR from the top of the leader board for money and/or points. But, we are suppose to take someones word for it that Archer is the number 2 player. Matchroom and Mosconi Cup and BCA leadership have all been lobbying for some ranking system to slip in on the billiard community and repeatedly there are folks trying to GAME the system. This post is just primer for the fargo system to be adopted. Its like putting lipstick on a pig. IMO

KD

Nothing is perfect. But if you look at the top USA list, and the top World list Mike Page has posted in the two threads, can you find more than a small handful of players you think are out of order? And if they are out of order, is it by a ton, or just a tiny amount?

Now, keep in mind, his system does this all automatically. No subjectivity. It only goes by hard data of match play.

 

[robsnotes4u]

It is SMOKE and MIRRORS!!!



Nothing has been provided regarding HOW the numbers are reached???



Is the US OPEN worth the same as the MEMPHIS OPEN???



One event is worth 20K and another worth 4k. The person with the most is the best!!!



Archer is FAR from the top of the leader board for money and/or points. But, we are suppose to take someones word for it that Archer is the number 2 player. Matchroom and Mosconi Cup and BCA leadership have all been lobbying for some ranking system to slip in on the billiard community and repeatedly there are folks trying to GAME the system. This post is just primer for the fargo system to be adopted. Its like putting lipstick on a pig. IMO



KD

Maybe I am reading your post wrong., if so, please disregard. As far a rating an individual why would a game between two individuals be rated any different because of location?


Sent from my iPhone using Tapatalk

 

[Kid Dynomite]

Nothing is perfect. But if you look at the top USA list, and the top World list Mike Page has posted in the two threads, can you find more than a small handful of players you think are out of order? And if they are out of order, is it by a ton, or just a tiny amount?

Now, keep in mind, his system does this all automatically. No subjectivity. It only goes by hard data of match play.

I am all for TRANSPARENCY this is not transparent???

Disclose what a TOP PLAYER must do to be #1, #2, #3, #4, #5???

old data is used to formulate this system is CRAZY. Archer 10 years ago is not Archer today!!!

KD