why the followers are more likely to lose

[Data]

That's kind of a bizarre example - wouldn't it be more accurate (or at least more representative) to assume that the inputs into the capper's decision were fixed throughout the year (he used his same method for devising his picks - not mixing inside information with mediocre information at random times)?

The inside information is an illustration of an edge based on certain conditions. The conditions could be of various nature. It could be pretty much anythig. How about two otherwise good players that seem equal but one dominates another? Not a lock but could be a good angle too. The capper has discovered those conditions before the others and that is where his edge is coming from.

[djiddish98]

Also, based on the theory, wouldn't the converse be true that it would likely be profitable to fade a handicapper after he's run hot, since we're assuming he's used up a portion of his +EV plays and has a higher likelihood of betting a -EV subset going forward?

Or am I misinterpreting something?

[Thremp]

Is it really necessary to use derogatory words? Here is how I see it. People tend to gravitate to those things that appear to be working better than others. For example, they see Mutual Fund A earning double digits while other Mutual Funds are earning single digits. What happens? More people begin to invest in Mutual Fund A. At this point, unless Mutual Fund A continues to produce at that higher rate or better, the new investor will earn less. Worse, once Mutual Fund A gets back to a more normal rate of return and the new investor is only earning single digits (or even losing money), the investor now sees another Mutual Fund that is earning double digits and the whole process starts over again. It is a well documented fact that the average investor does not get anywhere near the actual rate of returns of their investments. We can agree to disagree whether or not this scenario can be applied to a Handicapper A. I personally see the similarities. The choice to agree or disagree is yours. Joe.

This is clearly beyond your understanding (and likely anyone's since its bullshit).

[djiddish98]

The inside information is an illustration of an edge based on certain conditions. The conditions could be of various nature. It could be pretty much anythig. How about two otherwise good players that seem equal but one dominates another? Not a lock but could be a good angle too. The capper has discovered those conditions before the others and that is where his edge is coming from.

So in this example, the follower would enjoy the fruits of the capper but his edge would eventually fall to 0, as the others eventually figure out the edge.

I don't believe he would be more likely to lose though. His total winnings would be less than the capper, as the EV shrinks, but presumably, the capper doesn't make -EV bets if he knows his EV before betting. He just stops betting once the EV is 0, and the follower would do so as well.

[Data]

Sorry, this is what I was referencing. Is this an assumption for your theory (that's how I'm interpreting it).

The capper makes an asessment of the outcome probability (p) and compares them with the line (O - decimal odds). He knows his edge (E), with a certain confidence, of course, before the contest starts. E=p*O-1.

[djiddish98]

I would argue that this can be observable in the stock market, where the self-fulfilling prophecy of chartists does induce this sort of price movement, where the follower or retail investor is always the last to the party and the first to get burned.

However, I would agree with TC that these sort of oscillations are not as predictable in gambling markets when following handicappers.

I don't know how one would arrive at a confident assessment that a handicapper has essentially "used up" his edge as opposed to just hitting a lucky subset. Care to enlighten?

[CrimsonQueen]

I think it's obvious to say that overall followers are more likely to lose, without even arguing any details at all about following a winning capper.

Either 1 of 2 things is happening when you follow someone:
1. You're following a good winning capper.
2. You're following a bad capper who got lucky.

If 1, then forget it. You all can argue about that till you're blue in the faces. He's a winner, follower jumps on and off board...who cares.
If 2, then what happens is he probably happens to win 10 in a row or whatever the case may be, and people decide to follow him. Then maybe for the rest of the season he goes 2-9. Follower goes 2-9, where the losing capper actually goes 12-9.

So overall, there's no way that you can't say followers lose more than the cappers, because NO ONE sees a system going 0-14 and says "I'm gonna follow THAT system!!" and then it goes 6-2 til the end of the season, and the follower makes more than the capper.

[Data]

So in this example, the follower would enjoy the fruits of the capper but his edge would eventually fall to 0, as the others eventually figure out the edge.

Yes, if the follower stays with the capper but this does not happen.

I don't believe he would be more likely to lose though. His total winnings would be less than the capper

Assuming that the capper does not make -EV bets, the follower's record will converge at 50% and he will lose due to the juice. Also, that assumption is not neccessarily true as the winning capper may make -EV bets due to the juice and to selection bias.

[djiddish98]

Assuming that the capper does not make -EV bets, the follower's record will converge at 50% and he will lose due to the juice. Also, that assumption is not neccessarily true as the winning capper may make -EV bets due to the juice and to selection bias.

If the capper knows his edge with a certain confidence, why would he knowingly make -EV bets? Wouldn't he reach a point where his confidence was low enough to stop betting before he converged to 50%?

Why would the follower's record converge to 50% as opposed the breakeven point where EV = 0?

[Data]

If the capper knows his edge with a certain confidence, why would he knowingly make -EV bets??

He could be wrong.

Wouldn't he reach a point where his confidence was low enough to stop betting before he converged to 50%??

I am not talking about capper's confidence in his methods but about the confidence interval (look it up) of his estimates.

Why would the follower's record converge to 50% as opposed the breakeven point where EV = 0?

If his ATS and totals bets are 0EV then his record will converge to 50%.

[Peregrine Stoop]

I think it's obvious to say that overall followers are more likely to lose, without even arguing any details at all about following a winning capper. Either 1 of 2 things is happening when you follow someone: 1. You're following a good winning capper. 2. You're following a bad capper who got lucky. If 1, then forget it. You all can argue about that till you're blue in the faces. He's a winner, follower jumps on and off board...who cares. If 2, then what happens is he probably happens to win 10 in a row or whatever the case may be, and people decide to follow him. Then maybe for the rest of the season he goes 2-9. Follower goes 2-9, where the losing capper actually goes 12-9. So overall, there's no way that you can't say followers lose more than the cappers, because NO ONE sees a system going 0-14 and says "I'm gonna follow THAT system!!" and then it goes 6-2 til the end of the season, and the follower makes more than the capper.

like this post

along the same lines, followers lose because the leader either is 1) smarter than the market or 2) not, on each individual game. When it's scenario one, the market will move towards the leader's opinion. The follower will get a worse price and so on. When it's scenario two, the follower is making a -ev bet.

[djiddish98]

Wouldn't it converge to ~52.38%, assuming a -110 / -110 line (is that what we're assuming)?

Also, if his confidence interval starts to widen or shift where there's a sufficiently large probability of having a -EV bet, wouldn't that be another case to call it quits?

I'm basically arguing that if a capper reasonably knows his edge (a huge if), he'll bail with making picks before the follower loses money to the vig.

[Data]

I don't know how one would arrive at a confident assessment that a handicapper has essentially "used up" his edge as opposed to just hitting a lucky subset. Care to enlighten?

This has already been explained. If we assume that the capper is a winner, his EV per bet will vary. It is more likely that he played some of his higher EV bets during the winning streak.

[Data]

I'm basically arguing that if a capper reasonably knows his edge (a huge if), he'll bail with making picks before the follower loses money to the vig.

While replying to your last couple of messages I kept noticing that you are completely missing jumping-in-and-out on the follower's part.

[djiddish98]

While replying to your last couple of messages I kept noticing that you are completely missing jumping-in-and-out on the follower's part.

True - I have not been considering that.

So the irrationality of the follower could conceivably lead him to a point where he bails once his profit from following is < 0, which, due to variation in handicapping results, is likely to occur at some point in time while following.

I don't mean to discount this idea, but isn't this more a behavioral issue, as opposed to reflecting something related to the handercapper's underlying EV?

[tomcowley]

This has already been explained. If we assume that the capper is a winner, his EV per bet will vary. It is more likely that he played some of his higher EV bets during the winning streak.

Which does nothing to explain why he is not now expecting to have more high-EV bets this year than he expected to at the start of the year.

[Data]

So the irrationality of the follower could conceivably lead him to a point where he bails once his profit from following is < 0, which, due to variation in handicapping results, is likely to occur at some point in time while following.

Also, the follower may bail out with overall winning record after a series of losses.

I don't mean to discount this idea, but isn't this more a behavioral issue, as opposed to reflecting something related to the handercapper's underlying EV?

Yes, it is interesting from behavioral standpoint. We do have to analyse the capper to find a theoretical basis for the observed phenomena.

[Data]

Which does nothing to explain why he is not now expecting to have more high-EV bets this year than he expected to at the start of the year.

Where those high +EV opportunities are coming from? What's their nature?

[tomcowley]

Who knows? What is the distribution of H(50) before the season starts?

[Data]

Define a winning player.

[tomcowley]

Assuming we're talking about looking forward, and let's restrict it to one upcoming season for simplicity: A player whose methodology, applied to all (weighted) possiblities for the upcoming season, has a positive expectation.

In math terms, given a season possibility i, and game j in that season (out of N)

his estimate of his EV is Sum i=1..whatever sum j=1..n P(i)*H(ij) where H(ij)>0

When Sum i=1..whatever sum j=1..n P(i)*realev(ij) when H(ij)>0 is >0, he's a winning player.

[Data]

The following may use some defining/meaning clarifying:
possibility
season possibility
P
H

[tomcowley]

season possibility: a way the games (and all other information about the season) and lines offered can happen
P: probability, with the normal definition, and sum i=1..whatever P(i)=1.
H(ij): Handicapper's estimate of his EV in game j of season possibility i (0 if he thinks there's no edge/insufficient edge)

[HedgeHog]

Data is correct--"Followers" are likely to lose. Why?, because "Leaders" are likely to lose too. Tailing hot cappers and fading cold ones is a loser's fallacy. At least 98% of all bettors lose long term.

[Data]

realev?

More importantly, you do not know why the winning methodology comes up with H(ij)>0, it just does this in a black box manner, correct?

[tomcowley]

All probability statements you make are conditional on the information you know. It's often left unstated, but it's important. They're also conditional on how you interpret the information you have. The real EV would be the probability you would assign if you knew all possible information and interpreted it perfectly. Obviously you can never know that your statement of EV is the real EV even if it is.

A casino thinks its EV on a particular craps bet is +x. The player who snuck in loaded dice thinks it's -y (for the casino). In reality, he only slipped in one loaded die and one normal die and it's actually +z for the casino. Neither the casino's nor the cheater's statements of probability are "wrong" per se, based on the information they have, but neither describe reality. You, me, BtC, and a bunch of others have effectively identical line/score NFL DBs. He thinks his donk teasers are +EV. Pinnacle, Cris and everybody who could bet 1000 on alt -240 disagree. Obviously we consider our methodology far superior to his, based on superior (any) understanding of math, logic, and common sense, but it's impossible to prove with 100% confidence that we're correct.

So back to the question, we have to evaluate the capper's H(ij) against our own estimate of realev (a huge percentage of donk cappers think their bets are +ev (like BtC teasers) and are wrong with high statistical confidence after enough bets.. just because he thinks he's +EV doesn't mean he is). Knowledge of how H(ij) is calculated is not necessary, but information about it can be used (everything else being equal, stat-based capper > Paul the Octopus = BtC for example). In the case of a semi-donk follower who doesn't know anything about what the line should be (other than what it is), but has some knowledge of sports capping and isn't retarded, information about the capper's method can be used in his evaluation of how likely H(ij) is to be the real EV, or how likely he considers the claimed edge to actually be an edge.

[Thremp]

I question your lol anecdotal evidence.

I tailed a tout earlier tonight. I won 1-0. tule siia ja ime mu suurt muuni u fagg0t.

[Data]

This is not what I am trying to get from you. I presented a model of a winning player that you seem to disagree with. I am confident that my model is fairly good. However, at this point I am unsure if you critique comes from implying a better model and that hypothetical better model is what I am trying to get you to present. I want to assume that a capper is a winning player. Now, what makes him win? So far I got from you that a winning player is the player whose plays are +EV. That's cool but we kinda knew that.

[TomG]

[subs]

^^^ best post so far

[tomcowley]

This is not what I am trying to get from you. I presented a model of a winning player that you seem to disagree with. I am confident that my model is fairly good. However, at this point I am unsure if you critique comes from implying a better model and that hypothetical better model is what I am trying to get you to present. I want to assume that a capper is a winning player. Now, what makes him win? So far I got from you that a winning player is the player whose plays are +EV. That's cool but we kinda knew that.

What is your model of a winning player?

[Data]

What is your model of a winning player?

I have been referring to the cappers and the systems as possible leaders for a followers as they both can be viewed in a similar way as identifiers of uncommon conditions which historically have not been priced correctly. That is my model in a essense.

[tomcowley]

What causes those conditions to appear during the season if and when they appear?

[Data]

The causes are past performances and game day factors such as refs, injuries, weather and so on and so forth.

[tomcowley]

Are we discussing cappers who create their own estimate of a true line, or people who will bet the line at any price?