Morning Line and Final Odds Favorites

[OverlayHunter]

I'm guessing that a number of people on this site have some insights into the following.

So far in 2023, morning line favorites are winning about 33% of races and about 63% of the ml favorites are also the final odds favorites. The final odds favorites are winning about 38%.

If you deem a morning line favorite to be vulnerable, what should the overall winning percentage of those horses be to confirm that they actually - as a group - were vulnerable?

If you deem a morning line favorite to be false, what should the overall winning percentage of those horses be to confirm that they actually - as a group - were false?

If you deem a final odds favorite to be vulnerable, what should the overall winning percentage of those horses be to confirm that they actually - as a group - were vulnerable?

If you deem a final odds favorite to be false, what should the overall winning percentage of those horses be to confirm that they actually - as a group - were false?

Thank you for the assistance.

[MJC922]

IMO any overall win rate on that which you deem to be 'vulnerable' and which ends up winning less percentage-wise relative to the win pool's final totals, that's all clearly evidence in your favor.

If under 2-1 on the board is vulnerable in your mind and it wins 20% of the time or less IMO then you're looking at something very encouraging. Check my math but I believe even at 9/5 after take you're neutralizing takeout by maybe 10% given 20% win rate.

[OverlayHunter]

IMO any overall win rate on that which you deem to be 'vulnerable' and which ends up winning less percentage-wise relative to the win pool's final totals, that's all clearly evidence in your favor.

If under 2-1 on the board is vulnerable in your mind and it wins 20% of the time or less IMO then you're looking at something very encouraging. Check my math but I believe even at 9/5 after take you're neutralizing takeout by maybe 10% given 20% win rate.

MJC922, thank you for the reply.

I may be missing your point and may see with more clarity after I get more sleep, but I think your reply speaks more to overlays and underlays. I'm trying to establish a benchmark (or benchmarks) to use as a guide to determine if I can/should - more or less profitably - bet against a favorite.

In my head, if the set of ML favorites wins 33% of their races and the set of Final Odds favorites wins 38% of their races, there should be some percentage of those favorites that do better than 33%/38% because they and/or the other horses have important characteristics that make the favorite more likely to win. Conversely, there should be a combination of factors that make the favorite less likely to win. I'm not asking for any secret factors that cause the favorite to win more often or lose more often, I'm just trying to get a sense of what makes a race more worthwhile to spend time on or less worthwhile to work on based on the anticipated win percentages of helpful or hurtful factors that affect the favorite.

For example, perhaps a lone front running ML and Final Odds favorite running against a field of horses that have never been closer than 10 lengths at the stretch call. That front runner may lose but its chances of winning go up significantly more than the averages of 33%/38%. For the sake of argument, let's say those win 40%/47% or more. (The odds it goes off at will determine whether it is an overlay or not and whether it should be wagered or not.)

So I'm looking for a general guide to help determine the races I should more closely look at; that is, is the favorite likely to lose more than typical. If that is true, what are good benchmarks to use to call them vulnerable or false? If the ML favorite typically wins 33%, but some combination of factors causes them to win only 25% of the time I'm guessing they should be considered at least vulnerable. But this is the point of the thread, I don't have the knowledge to know what the benchmarks should be. How much should the win% decline before calling a favorite vulnerable or false in the way we use those terms.

[dnlgfnk]

Originally Posted by OverlayHunter

MJC922, thank you for the reply.

I may be missing your point and may see with more clarity after I get more sleep, but I think your reply speaks more to overlays and underlays. I'm trying to establish a benchmark (or benchmarks) to use as a guide to determine if I can/should - more or less profitably - bet against a favorite.

In my head, if the set of ML favorites wins 33% of their races and the set of Final Odds favorites wins 38% of their races, there should be some percentage of those favorites that do better than 33%/38% because they and/or the other horses have important characteristics that make the favorite more likely to win. Conversely, there should be a combination of factors that make the favorite less likely to win. I'm not asking for any secret factors that cause the favorite to win more often or lose more often, I'm just trying to get a sense of what makes a race more worthwhile to spend time on or less worthwhile to work on based on the anticipated win percentages of helpful or hurtful factors that affect the favorite.

For example, perhaps a lone front running ML and Final Odds favorite running against a field of horses that have never been closer than 10 lengths at the stretch call. That front runner may lose but its chances of winning go up significantly more than the averages of 33%/38%. For the sake of argument, let's say those win 40%/47% or more. (The odds it goes off at will determine whether it is an overlay or not and whether it should be wagered or not.)

So I'm looking for a general guide to help determine the races I should more closely look at; that is, is the favorite likely to lose more than typical. If that is true, what are good benchmarks to use to call them vulnerable or false? If the ML favorite typically wins 33%, but some combination of factors causes them to win only 25% of the time I'm guessing they should be considered at least vulnerable. But this is the point of the thread, I don't have the knowledge to know what the benchmarks should be. How much should the win% decline before calling a favorite vulnerable or false in the way we use those terms.

The only benchmark is to produce winning expectation. If that 38% is cut in half, you're redistributing 19 win pct. points to the remaining field. A 2-1 (26%?) second choice becomes 32%, and pays $6.00 thirty-two percent of the time for a 4% loss.

Cutting the 38% favorite to 10% redistributes twenty-eight win pct. points. That 2-1 (again, is the current win pct. for a 2-1 horse 26%? It used to be.) becomes a 36% winner at $6.00, = $216 and an 8% ROI.

I don't get involved in such hairsplitting. I merely use my factors to assign zero chance to horses I downgrade, and not necessarily the favorite. But from the above, one might judge that it requires a redistribution of roughly 22 win pct. points to break even.

[MJC922]

I think I understand what you're asking but it's kind of an individual thing I suppose. A working definition for a false favorite might be looked at in terms of the 50% overlay rule of thumb. By that I mean a morning line favorite could probably be considered 'false' if the odds are MLO x 1.5 (or more) in your mind. The actual tested win rate of such horses might then be considered to pass the 'test' at the midway point between the two, roughly MLO x 1.36 so 2 x 1.36 = 2.72 and converted to expected winners would be 1 divided by (2.72 + 1) = 27% or less, something along those lines might be what you're looking for? I've never seen it actually defined in print, only the 50% overlay bit which itself is just a rule of thumb probably intended to build in a sufficient cushion for error and / or late odds changes (kind of before drastic late odds changes at some tracks became the rule rather than the exception). Like I said kind of an individual thing to define IMO so if you want to set 'false' at MLO x 2 instead of 1.5 then have at it.

[OverlayHunter]

MJC922 and dnlgfnk thank you for both of your perspectives. I'll think about both replies and see if I can implement something meaningful and helpful from them.