FargoRate

[Poolplaya9]

Cleary and I have went countless rounds with Mike about this very thing and I think we all have to just to agree to disagree - at least for now till we see what happens later on.But yes, I totally agree with you that this is a major flaw and just can't seem to wrap my head around " it just doesn't matter " explanation .

Mike explained it in post #18 but I can totally see where lots of people won't be able to comprehend and understand how it is able to work. The best thing for someone in your position to do to put their mind at ease about it is to try to think of the people you know that play mostly on bar tables, and those that play mostly on 9 footers (or whatever other two table sizes). Look up their Fargo ratings and try to find guys from each group who have ratings that are very close to each other, like maybe you will find someone who is around a 620 speed from each of the groups, and you might also find two others who are about a 550 rating, one from each of the groups, etc. Then ask yourself if you believe they are actually pretty evenly skilled just like their ratings suggest. Ask yourself if they have fared about equally when they have faced each other in local tournaments, or when they have gambled together (they splits wins and often have tight matches etc). Ask yourself if say the 620 you can think of from each of the groups would be a close match if they were to play on say an 8ft table.

I think you will also find that the people who are rated the same, play about the same, even if each plays primarily on different sized equipment from the other. Mike has also found the same thing, over and over and over, because he isn't relying on a hope or a theory that the FargoRate system can deal with this issue, he has tested it out and proved it out repeatedly that in "real life" people that are rated the same end up being pretty evenly matched even when they primarily play on different size equipment and that FargoRate does in fact effectively deal with this issue.

 

[BmoreMoney]

Mike explained it in post #18 but I can totally see where lots of people won't be able to comprehend and understand how it is able to work. The best thing for someone in your position to do to put their mind at ease about it is to try to think of the people you know that play mostly on bar tables, and those that play mostly on 9 footers (or whatever other two table sizes). Look up their Fargo ratings and try to find guys from each group who have ratings that are very close to each other, like maybe you will find someone who is around a 620 speed from each of the groups, and you might also find two others who are about a 550 rating, one from each of the groups, etc. Then ask yourself if you believe they are actually pretty evenly skilled just like their ratings suggest. Ask yourself if they have fared about equally when they have faced each other in local tournaments, or when they have gambled together (they splits wins and often have tight matches etc). Ask yourself if say the 620 you can think of from each of the groups would be a close match if they were to play on say an 8ft table.

I think you will also find that the people who are rated the same, play about the same, even if each plays primarily on different sized equipment from the other. Mike has also found the same thing, over and over and over, because he isn't relying on a hope or a theory that the FargoRate system can deal with this issue, he has tested it out and proved it out repeatedly that in "real life" people that are rated the same end up being pretty evenly matched even when they primarily play on different size equipment and that FargoRate does in fact effectively deal with this issue.

Look, I'm not getting into algorithms, absolute VS relative or anything else like that. All I'm doing is using decades of real world experience to draw my opinion on. I've been in this situation , seen this situation countless times - player a and player b are matching up . Player a is spotting player b ( doesn't matter what the actual spot is but let's just say the 6 ball ). Player a is spotting player b the free 6 on a 9 footer. They decide to play next time on the barbox. Player b NEVER gets as much weight on the barbox as on the 9 foot, period. So, if everything is truly equal and it doesn't matter - why is this so? Common sense tells me there must be a difference for the spots being different between the 2 tables. No? As I said, either as a player, backer, side better or sweater, I have seen this scenario time and time again.

 

[hang-the-9]

Right on! LOLOLOLOL!



Cleary and I have went countless rounds with Mike about this very thing and I think we all have to just to agree to disagree - at least for now till we see what happens later on.But yes, I totally agree with you that this is a major flaw and just can't seem to wrap my head around " it just doesn't matter " explanation .

To have a good rating using statistics, you need to gather the data. If you do not have the data, there is no way to do it. If a bird species lived in a single tree on the planet it is not likely to be discovered, but is that the fault of the scientists not looking enough? Or just that the bird is to isolated? How can you possibly have a known rating in detail of any player that does not play outside of a small circle of people, and those people don't play anyone outside their circle either? The only way to rate anyone then is to see how well they play vs the ghost then compare that to another player that does as well that does have a Fargo rating. Or just use a known player's rating and just say if this 600 can run out a rack in 9 ball 50% of the time, and this player from the woods I have never seen can do the same, this other player is a 600.

 

[BmoreMoney]

To have a good rating using statistics, you need to gather the data. If you do not have the data, there is no way to do it. If a bird species lived in a single tree on the planet it is not likely to be discovered, but is that the fault of the scientists not looking enough? Or just that the bird is to isolated? How can you possibly have a known rating in detail of any player that does not play outside of a small circle of people, and those people don't play anyone outside their circle either? The only way to rate anyone then is to see how well they play vs the ghost then compare that to another player that does as well that does have a Fargo rating. Or just use a known player's rating and just say if this 600 can run out a rack in 9 ball 50% of the time, and this player from the woods I have never seen can do the same, this other player is a 600.

Well SOMEHOW people had figured out how to match up pretty Fargo since the inception of pool, so yes not only can it be done but has been forever just using your knowledge and eyes.

 

[Poolplaya9]

Look, I'm not getting into algorithms, absolute VS relative or anything else like that. All I'm doing is using decades of real world experience to draw my opinion on. I've been in this situation , seen this situation countless times - player a and player b are matching up . Player a is spotting player b ( doesn't matter what the actual spot is but let's just say the 6 ball ). Player a is spotting player b the free 6 on a 9 footer. They decide to play next time on the barbox. Player b NEVER gets as much weight on the barbox as on the 9 foot, period. So, if everything is truly equal and it doesn't matter - why is this so? Common sense tells me there must be a difference for the spots being different between the 2 tables. No? As I said, either as a player, backer, side better or sweater, I have seen this scenario time and time again.

Yes, a bar table doesn't show the differences between players as easily as a 9 ft table. But FargoRate is still able to make it all work out in the end and be pretty accurate even when they use info from all table sizes. Even if you aren't able to understand how, all that is really important in the end is if it is does it or not. Which is why I suggested looking at the real world examples instead of theorizing. Find the players who are rated the same but primarily compete on different equipment and see if they are in fact pretty evenly matched. Mike has done this repeatedly. Instead of relying on your theory for your belief, find out what the actual reality is. It's easy enough to do. The proof is in the pudding as they say.

 

[Poolplaya9]

To have a good rating using statistics, you need to gather the data. If you do not have the data, there is no way to do it. If a bird species lived in a single tree on the planet it is not likely to be discovered, but is that the fault of the scientists not looking enough? Or just that the bird is to isolated? How can you possibly have a known rating in detail of any player that does not play outside of a small circle of people, and those people don't play anyone outside their circle either? The only way to rate anyone then is to see how well they play vs the ghost then compare that to another player that does as well that does have a Fargo rating. Or just use a known player's rating and just say if this 600 can run out a rack in 9 ball 50% of the time, and this player from the woods I have never seen can do the same, this other player is a 600.

This is correct if everybody in each circle has never played outside their circle. But invariably there will be people in any circle who have played someone outside that circle. When that happens, you are able to make comparisons through indirect but accurate means. It is a very involved and complicated version of this: if Adam plays exactly even with Bob, and Bob plays exactly even with Carl, then you can infer that Adam plays exactly even with Carl even though Adam and Carl have never played each other, or even met, and may even live on opposite sides of the world.

FargoRate isn't usually having to rely on one single relation like this to judge a persons speed though and ultimately can compare everybody to everybody else that same way. It is able to compare everybody you have ever played, against everyone they have ever played, and on and on and you end up with many connections and comparisons that make it very accurate. It that respect it is kind of the same as they say with STD's where you aren't just sleeping with that one person, but with everyone they have ever slept with, and then everyone all those people have ever slept with, and so on.

 

[mikepage]

Look, I'm not getting into algorithms, absolute VS relative or anything else like that. All I'm doing is using decades of real world experience to draw my opinion on. I've been in this situation , seen this situation countless times - player a and player b are matching up . Player a is spotting player b ( doesn't matter what the actual spot is but let's just say the 6 ball ). Player a is spotting player b the free 6 on a 9 footer. They decide to play next time on the barbox. Player b NEVER gets as much weight on the barbox as on the 9 foot, period.

I agree the correct ball-spots will be different on the two tables

So, if everything is truly equal and it doesn't matter - why is this so?

There are a lot of things that are different playing on a 7-foot vs a 9-foot table. I don't hear anybody claiming everything is equal.

Common sense tells me there must be a difference for the spots being different between the 2 tables. No? As I said, either as a player, backer, side better or sweater, I have seen this scenario time and time again.

This is a ball-spot issue. You will never hear us say something like "a 50-point Fargo Rating gap is the 7-out." That's because there is no such correspondence. The 7-out is a large spot for weak players and a progressively smaller spot for better players.

But in general if you can give someone 5 games on the wire in a race to 20 on a 9-foot table, that's also about what you can give them on a 7-foot table (barring a situation where one of them is a fish out of water on one of the tables)

 

[mikepage]

[...]Common sense tells me there must be a difference for the spots being different between the 2 tables. No? As I said, either as a player, backer, side better or sweater, I have seen this scenario time and time again.

Here is a way to think about it.

Imagine two players who are 50 points apart in Fargo Rating, a 450 and a 500.
And they have a certain ball spot that works, say the 8-ball

Now imagine these two players start magically improving--but at the same rate.

After one week they are 550 and 600
then after the next week they are 650 and 700

The gap in speed stays the same.
The ball-weight spot is going to have to go up. A 650 has no change getting the 8 against a 700. I think we all agree on this point.

Now imagine YOUR players a and b have a certain spot on the 9-foot table. When they move to the bar box, they both in effect get better. As the equipment gets easier both players in essence get better--and the spot needs to go up for the same reason it did in the first scenario.

 

[Oze147]

If fargo rating is accurate, at least to a certain level, I wonder if someone has ever used it for betting.

If this system really allows to compare different players all over the world, it should be possible to predict the winner of a tournament, a league etc.

So if player X has a fargo rate of 700 he has a 90% chance to win the MadeupClassic 2017, because all other players in the draw are ranked below.
But if he enters Dreamland Open, his chances drop to 40% percent, because most of the other entrants are ranked higher...or something like that.

...but since I didn't hear about big betting wins, either I have missunderstood this system, or it isn't that compareable as some people say it is.

 

[BeiberLvr]

Now imagine YOUR players a and b have a certain spot on the 9-foot table. When they move to the bar box, they both in effect get better. As the equipment gets easier both players in essence get better--and the spot needs to go up for the same reason it did in the first scenario.

You know this isn't totally true.

Some players actually play worse when moving over to a bar table.

 

[iusedtoberich]

To Beiber:

No one in the world plays worse on the bar table! Where do you come up with that?!

Maybe if it was a pos table full of throw up on it from the bar drunks, rolling off because naked drunk girls were dancing on it, and rotted rails from all the spilled alcohol!

No way otherwise.

 

[hang-the-9]

You know this isn't totally true.

Some players actually play worse when moving over to a bar table.

It is impossible to track statistical anomalies. Fargo is the overall average of all results, a few discrepancies will always happen. You may as well start talking about Fargo rating of someone when they play in 110 degree weather vs 80 or when they are tired or when they forgot their cue at home or when they are playing their pool idol and are more nervous, or anything that you can't predict and track well. In that case, feel free to adjust the Fargo rating you use for a bet all you want with someone. You can say, well normally I would give you even money since they are both rated 600 but this guy can't play well on a Sunday after a birthday party on Saturday and before a holiday off work on Monday so I need a game spot.

It is impossible to track that statistic since it's not really a statistic.

 

[jimmyg]

While rating players in this fashion has some merit, without players actually competing against each other on a regular basis any sport wide ratings system will be seriously lacking, probably limited to the same individual statistics that can be compiled by players simply playing the ghost.

 

[Poolplaya9]

If fargo rating is accurate, at least to a certain level, I wonder if someone has ever used it for betting.

If this system really allows to compare different players all over the world, it should be possible to predict the winner of a tournament, a league etc.

So if player X has a fargo rate of 700 he has a 90% chance to win the MadeupClassic 2017, because all other players in the draw are ranked below.
But if he enters Dreamland Open, his chances drop to 40% percent, because most of the other entrants are ranked higher...or something like that.

...but since I didn't hear about big betting wins, either I have missunderstood this system, or it isn't that compareable as some people say it is.

FargoRate can be used to do exactly what you are talking about, calculate the odds of somebody winning an event, but it would take some work and math knowledge on your end to be able to calculate it out. It gives you everything you need though.

One thing you have to keep in mind though is that a for a player to be favored to win an event (favored meaning more than 50% likely to win the whole thing), they have to not just be better than everyone else in the event, but significantly better. It would be like taking one person against the field in a bet and your odds for winning on that one player against the field aren't very good unless your player is not just better than everybody else, but a lot better, because there are usually a number of upsets that happen along the way in any event. For example, let's say you have a guy who is better than anyone else in a 64 man event, and in fact is favored to win every single match 2/3rds (66.66%) of the time, which means that he is a pretty dominant player who is significantly better than everybody else. Well that guy still isn't favored to win the whole event. He will definitely have a better chance to win it than anybody else does, but his odds for winning are still less than 50% and you can easily calculate out what his exact odds are if you wanted to and know how to do it, all with the info provided by FargoRate.

If you go to http://fairmatch.fargorate.com/ you can look up the rating for any rated player by going to "find a player", and you can look up the odds/chances for either player winning any single match up by going to "find match odds". You can even find out how many games on the wire need to be spotted to make for an even match between two players of different skill levels by going to "find a fair match". And yes FargoRate can and has been used for betting, at least by smarter players. It is awesome info and when used along with your own good knowledge (who doesn't tend to play well out of the country, who has the flu right now, who seems to rise to the occasion or fall apart under the biggest pressure, who seems to have who's number when they match up even though they play the same speed, etc) it would usually give you an advantage over someone else who is relying only on their subjective judgement alone.

 

[hang-the-9]

While rating players in this fashion has some merit, without players actually competing against each other on a regular basis any sport wide ratings system will be seriously lacking, probably limited to the same individual statistics that can be compiled by players simply playing the ghost.

Well, at some point you will have a web of players ranked together even if two specific players never played each-other. If I play player A 10 times and beat them 8 times, and player A has played player B 10 times and beat them 5 times, I have a pretty good shot at beating player B at about the same rate I beat Player A. Fargo can then infer that if I was a 600, and player A was a 540, then player B would be about a 540 also.

It does take a good data set to gather the info, but the more data they get, the more accurate the ratings will be and the greater number of players you have never met will have accurate ratings against you due to them at some point playing other players you played.

Say you sit in your home pool room and never leave. A guy you play with a lot and are even with travels to SBE and plays a bunch of matches. During those matches he wins, he loses, the players he played against have Fargo ratings. You now have a rating against those players as well since you play same speed as your buddy that played them. There is a connection between you, your buddy, the players your buddy played and the players they also have played. After a bit of this, the rating gets more and more accurate as it will even out one-off results like someone being on that day and beating a better player.

If you want to see how accurate the rating is, the Fargo ratings predict how close a match would be between two players, track some tournaments and see if that is fairly accurate. I have seen it be very accurate, even in one pocket. If it states a 700 will beat a 600 on average 7-3 in a race to 7, see how close that gets to tournament results.

 

[Poolplaya9]

While rating players in this fashion has some merit, without players actually competing against each other on a regular basis any sport wide ratings system will be seriously lacking, probably limited to the same individual statistics that can be compiled by players simply playing the ghost.

This isn't exactly true. If Adam has been playing Bob recently and they play dead even, and Bob has been playing Carl recently and they play dead even, then you know that Adam and Carl also play about even right now, even though they haven't played recently, and even if they have never even played at all. You don't have to be playing everybody else regularly to accurately establish where you are in the pecking order with everybody else, you only have to be playing some people. FargoRate compares everybody to everybody else through all these many connections and where anyone stands in the pecking order with everybody else can in this way be accurately established.

 

[pt109]

-----------------------------------------------------------
Six degrees of separation is the idea that all living things and everything else in the world are six or fewer steps away from each other so that a chain of "a friend of a friend" statements can be made to connect any two people in a maximum of six steps.
---------------------------------------------------------------
I feel, the more robustness, the better FargoRate is going to be.

 

[mikepage]

You know this isn't totally true.

Some players actually play worse when moving over to a bar table.

OK. But we should be careful about language. Sometimes when we say a player plays worse on the bar table, we mean he doesn't keep up with the same players he keeps up with on the 9-foot table.

It could be, though, that he is more likely to get out from the 5-ball on the bar table than on the 9-foot table. Maybe his sparring partner runs out 20% more moving from 9-foot to 7-foot and he runs out only 10% more.

 

[jimmyg]

While rating players in this fashion has some merit, without players actually competing against each other on a regular basis any sport wide ratings system will be seriously lacking, probably limited to the same individual statistics that can be compiled by players simply playing the ghost.

Just to clarify a bit: I'm not suggesting that conclusions cannot be drawn and, to some degree, be somewhat valid by this system...it is certainly a valid, mathematical, systematic, "rating system".

What I am suggesting is that without a more comprehensive and organized national/international player "tour", where players continually compete against one another, and points are awarded by placement, we're still in the middle of "a work in progress".

 

[mikepage]

If fargo rating is accurate, at least to a certain level, I wonder if someone has ever used it for betting.

Yes of course. People do that all the time. It is major new information for the oddsmakers too.

If this system really allows to compare different players all over the world, it should be possible to predict the winner of a tournament, a league etc.

Well yeah... But this is true in the same sense that you can predict whether you or the dealer will bust on the next card..... How's that one working out?

So if player X has a fargo rate of 700 he has a 90% chance to win the MadeupClassic 2017, because all other players in the draw are ranked below.
But if he enters Dreamland Open, his chances drop to 40% percent, because most of the other entrants are ranked higher...or something like that.

Yes

...but since I didn't hear about big betting wins, either I have missunderstood this system, or it isn't that compareable as some people say it is.

I'm not sure your standard is a reasonable one.

The Turning Stone Classic started yesterday. About three quarters of the players have established Fargo Ratings, which means about half the first-round matched will have BOTH players with established ratings. In fact of the 62 matches played, 31 of them had both established players. Here are the results. In 26 of the 31 matches the player with the higher Fargo Rating won the match. Of the 5 upsets, three were 9 to 8 scores.