Opportunities elsewhere

[Cesario!]

NOTE: I don't want to talk about JCapper, which is why this is in a new thread. More want to talk theorectically about the conception of shifting opportunities.

This whole JCapper thing got my thinking. I think the argument that seems to be missing (and that I think NJCurve might be suggesting...although I'm not aligning myself with anyone in this) is:

In a pari-mutuel system, when the odds shift towards one interest for any reason, it creates opportunities elsewhere. So, if for some reason, the odds move to make something 4-1 from 12-1, that additional money will create higher odds on other horses, perhaps making them overlays. Thus, the astute player doesn't care which horse he/she has rated ultimately gets the bet, so long as it presents value. The non-winner is the one who doesn't adjust their methods as the public (or "private public") catches up.



I don't think this is a controversial statement. Is it?

[sjk]

Not necessarily the case.

In a given race there may be 1-2 horses that are overlays (always in the eyes of the player). In a short field there may be none.

If there is extra play on the apparent value horses (due to the track take), the increase in prices on the other horses is not usually enough to provide the required value/edge to make them a play.

For the small/medium tracks it doesn't take a huge amount of money to make the value horses unplayable with no alternatives served up.

If there are additional randomizing factors (new surfaces) or well-bankrolled players with the same software you are using it is likely that the number of playable races would reduce considerably.

[Cesario!]

Not necessarily the case.

In a given race there may be 1-2 horses that are overlays (always in the eyes of the player). In a short field there may be none.

If there is extra play on the apparent value horses (due to the track take), the increase in prices on the other horses is not usually enough to provide the required value/edge to make them a play.

For the small/medium tracks it doesn't take a huge amount of money to make the value horses unplayable with no alternatives served up.

If there are additional randomizing factors (new surfaces) or well-bankrolled players with the same software you are using it is likely that the number of playable races would reduce considerably.

I do agree that track takes raises the necessary error that the public must make from "true" probabilities. But isn't this a baseline thing that affects ALL of the horses as a unit and the necessary difference that you need to see from your estimated probability?

In a "wiseguy" situation, the lowering of the odds on one horse necessarily raise the odds on the others. It's a proportional relationship.

Tell me what I'm missing.

[ryesteve]

In a "wiseguy" situation, the lowering of the odds on one horse necessarily raise the odds on the others. It's a proportional relationship.

Tell me what I'm missing.
The problem is that the other odds don't necessarily raise by enough. Let's say you like to bet horses that look ugly on paper, so the public shies away from them, but your methodology is able to see their hidden merit. Your hit rate's not great, but the average payout is around 8-1, so you're well into the black. Let's say for whatever reason, these horses you used to get 8-1 on are now going off at 4-1 or 5-1. At those odds, you can't really make money off them... but that kind of a shift doesn't really cause significant corresponding increases in the odds of the other horses... not enough to allow you to confidently identify any overlays there.

It's all a matter of what the magnitude of the betdown is. If you assume that these horses are instead bet all the way down to 6/5, then yes, it'd be hard not to see value elsewhere. But that scenario is far more unlikely.