BCA Nationals 8 Ball - Fargo Discrepancy

Much of the scrutiny Fargo is under is directly due to the Fan boys as well as Mark Grifin himself. When you tout something as the greatest thing to happen to pool you're setting expectations very high. Then, when questions/concerns/problems arise now Fan boys wanna say "give it time", or "its in its infancy", or "it's better than it used to be". We were told its the greatest thing to happen to pool, not that it'd simply be better than it was. And then, expect leniency on what they painted as the greatest thing since sliced bread. Fan Boys made their bed, now get to lay in it.



According to the website
109,427 players
3,466,280 games
130 countries

And 6 years

Not in its infancy. It doesn't need time.

Just because you don't like the answer, doesn't mean it isn't correct


Sent from my iPhone using Tapatalk
 
[...] I don't think 27 out of 32 top finishers playing in the wrong division is "a few." [...]

We have analysis coming, and first up will be an analysis or who is the group of players who came in with a 525 starter-rating and no other information. How did they do as a group? What is our best guess of the over/under for their actual ratings? I will say right now that the answer is under 500. So if you went to your table for your first match and your opponent was 525 with no games, the chances appear to be better than 50% that player actually plays below 500 speed.

I'm not sure where the above statement comes from. But I will say if you take ANY tournament, no matter how even the field, and then you try to compute ratings just based on that tournament, the people who finish high in the tournament will appear to be WAY UP, well above their actual speed. This is a selection bias. It is a little like looking at the lottery ticket in your hand, computing just how improbably it was YOU had that exact number, and then being amazed you have it.

Consider this: Imagine a 512-player single elimination tournament where every player has the EXACT SAME SKILL. This is not hard to find. All we have to do is hold a coin-flipping tournament, with good coins, where WE KNOW everybody is the same. So everybody flips coins...first round, second round... At the end there is a winner,

one player won 9 flips in a row, has a 9 and 0 record
one who is 8 and 1
two who are 7 and 1

Someone coming considering there might be biased coins might choose to look at these four players and say look at how extraordinary their records are. The coins must be strongly biased.

But wait! Even with no bias to the coins, there WILL be a winner, and there WILL be a second place.

We must be careful we are not amazed by our own losing lottery ticket...
 
We have analysis coming, and first up will be an analysis or who is the group of players who came in with a 525 starter-rating and no other information. How did they do as a group? What is our best guess of the over/under for their actual ratings? I will say right now that the answer is under 500. So if you went to your table for your first match and your opponent was 525 with no games, the chances appear to be better than 50% that player actually plays below 500 speed.

I'm not sure where the above statement comes from. But I will say if you take ANY tournament, no matter how even the field, and then you try to compute ratings just based on that tournament, the people who finish high in the tournament will appear to be WAY UP, well above their actual speed. This is a selection bias. It is a little like looking at the lottery ticket in your hand, computing just how improbably it was YOU had that exact number, and then being amazed you have it.

Consider this: Imagine a 512-player single elimination tournament where every player has the EXACT SAME SKILL. This is not hard to find. All we have to do is hold a coin-flipping tournament, with good coins, where WE KNOW everybody is the same. So everybody flips coins...first round, second round... At the end there is a winner,

one player won 9 flips in a row, has a 9 and 0 record
one who is 8 and 1
two who are 7 and 1

Someone coming considering there might be biased coins might choose to look at these four players and say look at how extraordinary their records are. The coins must be strongly biased.

But wait! Even with no bias to the coins, there WILL be a winner, and there WILL be a second place.

We must be careful we are not amazed by our own losing lottery ticket...
I would have to go back and find the quote but when the explanation by CSI was given why one bracket had 56% of the field in it, it was "because the people that were rated 525 were most likely 50 points too low anyways". I posted than that made no sense cause it was CSI that gave them the 525 starter rating. As a poster here basically said, they should have ranked them with a starter rating that put them higher up in the grouping. There is only one reason why they weren't. Also like the APA I believe never would start a male player as a 2, no male player should be given a starter rating of 400.
 
...Also like the APA I believe never would start a male player as a 2, no male player should be given a starter rating of 400.

In a day and age where we have the first woman presidential candidate nominated by the Democratic Party, one should be careful about sexism statements. :embarrassed2:
 
I would have to go back and find the quote but when the explanation by CSI was given why one bracket had 56% of the field in it, it was "because the people that were rated 525 were most likely 50 points too low anyways". I posted than that made no sense cause it was CSI that gave them the 525 starter rating. As a poster here basically said, they should have ranked them with a starter rating that put them higher up in the grouping. There is only one reason why they weren't. Also like the APA I believe never would start a male player as a 2, no male player should be given a starter rating of 400.



Curious.

How many games do you have in FargoRate? What is your rating? How long have you experienced it in your league?

I read such things as, "I would never lose to a 400", and "no male should be given a starter rating of 400".

I would guess a lot of people making such statements do not know what a 400 level player is.




Sent from my iPad using Tapatalk
 
Curious.

How many games do you have in FargoRate? What is your rating? How long have you experienced it in your league?

I read such things as, "I would never lose to a 400", and "no male should be given a starter rating of 400".

I would guess a lot of people making such statements do not know what a 400 level player is.




Sent from my iPad using Tapatalk


Using the letter grade system of D to Pro, what is a 400, 525, 600 level player in letters ?
 
Using the letter grade system of D to Pro, what is a 400, 525, 600 level player in letters ?

It strongly depends on where you are. In the northeast, that might be D, C, B.

In Minnesota it is C, B/A, and A/AA

There is no universal letter system.
 
According to the website
109,427 players
3,466,280 games
130 countries

And 6 years

Not in its infancy. It doesn't need time.

Just because you don't like the answer, doesn't mean it isn't correct


Sent from my iPhone using Tapatalk

Just because you think it's perfect, doesn't mean it is.
 
In a day and age where we have the first woman presidential candidate nominated by the Democratic Party, one should be careful about sexism statements. :embarrassed2:

What I find strange is I saw someone post a while back what the bronze, silver, gold, platinum number range was for bcapl and women and men had different numbers. I don't get that. A female 500 should be the same as a male 500.

It might have been a misprint or bad info, it wasn't from their site. I just found it strange.
 
Last edited:
According to the website
109,427 players
3,466,280 games
130 countries

And 6 years

Not in its infancy. It doesn't need time.

Just because you don't like the answer, doesn't mean it isn't correct


Sent from my iPhone using Tapatalk

Funny thing is, Mark Griffin said its in its infancy only a few posts up from mine.

Those aren't my words, those are the words of Mr Griffin & many Fan Boys. You may think it's the answer & it may be, for you. Doesn't mean it is for everyone.
 
Why wouldn't mini's be a good judge of speed? FargoRate essentially measures a person's average speed, which is overall what is most useful and accurate. To get a person's true average, you have to count all their games you are aware of. If you pick and choose which games and matches to count it is no longer a true representation of their avera

It did, you just didn't understand it I guess. No problem, you should have just said so. I will try to explain it a little more.

Your question was essentially that you couldn't think of any possible reason to explain how after the team event your FargoRating could have gone up while a teammates went down. Post #74 gave you two possible reasons.

One of the reasons that could explain this that was in that post is that FargoRate doesn't rate you based on whether you win or lose your matches against opponents, and it seems that you have the impression that it does. Whether you win or lose the match is immaterial. It only rates you based on how many games you won, compared to how many games you were expected to win based on your's and your opponent's ratings.

For example, lets say you play Shane Van Boening a race to nine. According to your current rating level (697) and Shane's current rating level (822) you are only supposed to get to 4 games by the time Shane wins the match. But lets say you lose 8-5. Well you won more games than a person rated a 697 is supposed to (you were only supposed to win 4 on average), so your rating is going to go up. It will probably only go up a minuscule amount since you didn't perform that much better than expected but your rating is going to go up none the less (although it may only up say .2 points, which after you round off you are still a 697).

Now say you lose to Shane 9-8. Now you won twice as many games as a 697 is supposed to win against Shane (you won 8 games but were only supposed to win 4) so your rating will go up even more than if you had lost 8-4 like in the first example. This time your rating might go up .8 points, and after you round off you are now a 698.

Now lets say you were to lose that match to Shane 8-2. Well then your rating would go down because you did not win the 4 games that were expected of a 697 speed player.

Now lets say you actually beat Shane in the match, and the final score is you won 9-6. Well you rating will go up even more than in any of the other examples. It isn't because you won that match though. That part doesn't matter. It is because you even further exceeded what you were supposed to do on average, which was only win 4 games. So you go up when you do better than a guy your speed was supposed to do, and you go down when you under perform how you were supposed to do against your opponent. So whether you actually won or lost a match is immaterial, it is by how much you exceeded or fell short of how a 697 would have done on average that makes the difference.

Something to keep in mind is that the more games you have in the system, the less difference any one match makes. If you have only ten games in the system and are a 697, and you lose 9-8 to Shane, your rating is going to go up, and go up a lot. But if you have a thousand games in the system your rating will only go up very slightly. Same thing when you perform lower than your rating.

When it comes to matches where you only won one game against an opponent (as in the team play at nationals) it gets kind of complicated to explain, but it essentially works the same way. You are generally going to go up when you win that one game against the guy, and how much you go up depends on how good he was. Since it was only one game it will generally only be a minuscule change in rating, again, depending on how many games you have in the system, but it probably won't even be enough to change your rating after you round off. It is the reverse if you lose the one game. All of the above (that was in post #74 just not spelled out nearly as detailed) leads to an answer to your original question of how after the team event you could go up and your teammate could go down, and one of those ways is because you could have performed better than you were expected to for a 697 (maybe you beat several 750 players, or maybe you beat all the players rated around your speed instead of just half of them, etc). That would be one explanation for you going up. And maybe your teammate perform slightly worse than he was expected to do for someone of his rating against the level of opponents that he played and therefore he went down.

Another reason that was in post #74 that could explain how you could go up but your teammate could go down is that you are re-rated every single day and your rating can change even if you have played no new games in the system. Basically if the ratings of the past opponents you have played against ends up going down on average, then your rating is going to go down too. And if the your past opponent's average ratings go up, then yours is going to go up too. And if they went up a lot, then yours will go up a lot.

What could easily have happened is that you beat a lot of players who were very under rated because they had no games in the system, or few games in the system. Lets say one of the guys you beat had a starter rating of 525, but as he played more matches and more singles and more mini's his rating went up to a 610 to closer reflect his true speed. Well part of your rating was taking into consideration that you beat this "525", but now that FargoRate knows that he is actually a 610, your win against a 610 is more "impressive" and means more or counts for more than when it was a win against a 525 and so your rating is going to go up to. And if you beat several of these significantly under rated players during the nationals whose ratings ended up going up quite a bit after they got more games in the system then your rating could see a pretty significant increase too. And being that there were lots of people with starter ratings, this is very likely a part of why you went up so much. Maybe your teammate didn't face as many of these unrated players, or if he did half of them were under rated but half of them were over rated so even after their ratings changed after they got more games in the system your teammate ended up staying about the same.

This was not mentioned in post 74, but while we are at it another reason you could have gone up while you teammate went down is that you could have won more games that he did during the team matches. How many games you won and how high your opponents are rated both make a significant difference to your rating.

And then as was already mentioned some time after you had asked the original question, the mini tournament results (where you did really well against high quality opponents) had already been entered and that certainly had a large affect on your rating too. In the mini's you way outplayed your previous rating and so your rating went up.

Thanks for the response.
 
What I find strange is I saw someone post a while back what the bronze, silver, gold, platinum number range was for bcapl and women and men had different numbers. I don't get that. A female 500 should be the same as a male 500.

It might have been a misprint or bad info, it wasn't from their site. I just found it strange.



Yes a 500 women and man would be the same.

I don't know what you are saying, but I assume you mean the division cut off numbers were not the same for men and women.

Fargo Billiards puts on tournaments named Quadrangles. Each time the cut offs are different, depending on the ratings of players. In other words Gold doesn't mean 425 - 525 for example. Gold is just a name.

I am not talking for CSI but this maybe the case. The splits between the two tournaments would be different because the two fields do not mirror each other in number and ratings.


Sent from my iPad using Tapatalk
 
What I find strange is I saw someone post a while back what the bronze, silver, gold, platinum number range was for bcapl and women and men had different numbers. I don't get that. A female 500 should be the same as a male 500.

It might have been a misprint or bad info, it wasn't from their site. I just found it strange.

A female 500 IS the same as a male 500, and when a female and male play in the mixed division, there is no issue.

But when we apply the word "platinum" to a female-only group, and the word means the top xx% of that group, it is going to be associated with a different rating range than if we use the same criterion for a different group.
 
Funny thing is, Mark Griffin said its in its infancy only a few posts up from mine.



Those aren't my words, those are the words of Mr Griffin & many Fan Boys. You may think it's the answer & it may be, for you. Doesn't mean it is for everyone.



The relationship between CSI and FargoRate might be considered in its infancy.

FargoRate itself is not when you look at the games, years, players, and countries.

I look as these as two separate things.




Sent from my iPad using Tapatalk
 
You know guys, it seems no one wants to answer the simple question I asked in other threads about Fargo. What happens to a player stuck by ability in the middle of a bracket? Any bracket. How does that player get a REASONABLE chance to drop out of that bracket other than to SANDBAG? Some say get better. What if your closer to the end of life than the beginning? What if the player is already playing as hard as they can? How long will that player continue to support the event with no chance to win a match or two let alone cash? Think everyone is stuck worrying about how their rating is determined and how it is determined in relation to other players. It's such a simple question. Why can't the powers that be answer it? I know, just wait and the answer will magicly appear. You'll see. Just wait a few more years.

Lyn
 
Funny thing is, Mark Griffin said its in its infancy only a few posts up from mine.
[...]

Is FargoRate in its infancy or is it honed and well developed?

The answer is yes.

We worked many thousands of person-hours for several years collecting data and building the infrastructure before anybody knew what was going on and before we hooked up with CSI. By the time we partnered with CSI a little over a year ago, we already had over a million games in the system, tens of thousands of players, many countries, and the full functioning global optimization. It was very tempting during those years to spill the beans and let people know what was going on but we didn't.

It was as though nobody had heard of the concept of a railroad, and we were secretly building trans-continental railroad tracks without every showing a train.

In the short time we have been working with CSI, our database has grown from 1.0 million games to 3.5 million games and we are only now introducing the LMS league management software that will bring games automatically in.

So think about this. Five years to get to 1.0 million games, one more year to get to 3.5 million games right now Things are growing very fast and the pace is only getting faster. We are at a VERY exciting time for CSI and FargoRate.

So back to the railroad analogy. Though we have tracks across the country with functioning trains, the North Dakota Soybean farmer may not be able get his crop to Asia until some of the side-routes are constructed.

So in terms of bringing this new concept of railroad transportation to the world, we are fully established. In terms of the typical farmer using rail traffic with his crops, we are in the infancy.

Some areas are quite established; other areas are quite thin. But the infrastructure for ALL areas is in place, and things are moving fast in an exciting direction.
 
You know guys, it seems no one wants to answer the simple question I asked in other threads about Fargo. What happens to a player stuck by ability in the middle of a bracket? Any bracket. How does that player get a REASONABLE chance to drop out of that bracket other than to SANDBAG? Some say get better. What if your closer to the end of life than the beginning? What if the player is already playing as hard as they can? How long will that player continue to support the event with no chance to win a match or two let alone cash? [...]

Lyn -- I recall responding to this in some more detail.

If, in fact, a player is in a situation where he has "no chance to win a match or two let alone cash," that player will MOST DEFINITELY be in a lower division.

But if that player is someone like you, who legitimately plays--currently--at 655 speed, who played 11 matches this year in Platinum singles divisions, winning 7 matches and losing 4 matches, then it would be unfair to the players in the Gold division to have the player added.

Surely you can appreciate this.
 
FargoRate

In ten years, the data collected and the data available will be much more than now. FargoRate will be much more common and be used in ways we can't even imagine.

Mikes railroad analogy is accurate. Improvements will be constantly made and the product will improve. That is what I was referring to as 'in its infancy'

Several have misinterpreted my statement. I did not say the method was incorrect or data was inaccurate-I was saying improvements in applications of this data will come over the years.
Some of these are a obvious (use it as a handicapping tool)-others aren't so obvious. The major area that will increase is data collection.

I never implied FargoRate was perfect. But more data will improve the accuracy of the system.

Hope this explains it more clearly.

Mark Griffin
 
Back
Top