Quote:
Originally Posted by o_crunk
From what I've seen, it doesn't work like "they are focusing on 2 or 3 live ones". They are betting all of them to a degree. I had 5,000+ races of tote hub data. There were just a handful out of that dataset where the total Elite return was $0 when they were in the win pool, literally maybe 20 races out of 5,000. In looking at what I had - what the total was they wagered in the pool and the total return and the matching it up with the win payoffs, it suggested they are betting all of them to different degrees, probably for different reasons if we were to look at the exposure in the other pools. They never clobber a win pool. Only bet above 20% of the entire pool 23 times out of 5,523 races. The median ROI in the win pool over those races was -0.1256 on the dollar. More than half the time they are losing quite a bit in that pool before the rebate. Only 578 races out of that sample did they double their money in the win pool.
But even if we take the lower level view, it still doesn't add up, particularly when the more egregious ones are taken into account. These are your 8/5 to 3/5 droppers that people really complain about. That would mean every single last dollar from the teams is going in one direction to one horse. They don't bet like that.
But maybe things have changed since I last looked. This was in 2019/2020. I don't discount that possibility since change is constant, especially for winners.
What seems most plausible to me is that covid brought even more money online to ADWs and less from the track. Someone earlier in the thread was saying 1/3rd is ADW. It's way more than that now. Just at the Oregon hub, 2/3rds of every dollar bet on US racing is through there and that hub doesn't include *all* the ADW handle. And as we've seen with some of the more recent ADW stuff (FD all/all supers, FD serving up 70 cent P5s instead of the 50 cent base increment, etc), I don't discount the ADW adopting the Robinhood model of selling their order flow, meaning they are holding their customers bets and dropping them at the host at the off. And now that everything is online, everyone is waiting to bet at the last minute. It's hard to imagine there used to be a world where you had to stand in a line for a teller to make sure you didn't get shut out. Now, you and literally every one else betting a race can wait, even with the video lag, and make your bets within the last 30 seconds before off remotely. And depending on who you bet with, that money is definitely not hitting the pool until it has already closed.
Any way you slice, the teams can't be the only reason for the odds drops. They don't bet enough and even if we are taking the word of Dan Ross in the article that they are consistently winning before the rebate (which I don't, I suspect this tidbit comes from Pat Cummings who was tweeting this same FBP stuff 6 months ago), there is definitely no reason for them to be in the win pool at that point and slamming horses. The win pool is a loss leader / volume builder for them, gets them the rebate. They get nothing out of making -EV bets even if they win.
|
Fundamentally, with half the field being overlays to various degrees and half the field underlays to various degrees once we apply takeout to that 'reality' there can't possibly be more than between 1 to 3 overlays in the average race, can we agree on that? Frankly the third is probably breakeven or negligible, right? I mean that's just logic.
I'm not saying
some races here and there don't have more than 1-3, certainly in the derby with a huge field there will be more than 3 but in the average race. Sure, I can walk up in an 8 horse field and mattress mac a 90-1 shot so then we will have 7 overlays, let's nevermind that, let's talk about the average race with the 1-3 overlays.
What could be done is to dutch those overlays and then turn around and give back those calculated profits by hedging some of the underlays to various degrees to the point where the end result is no longer a profit on the overlays but a break even on the group. Then the rebate check arrives at the end of the month and makes it all good.
That's the only way I can possibly see the scenario occurring as you describe it where they demonstrate cashing every race no matter who wins. The other pools are going to be played the same way, they know what the percentages look like for every horse to finish underneath in every slot. That would come from a very large historical DB.
