Low ROI first time starter trainers idea.

[teddy]

When you have a race with two favorites or two first time starters by the big name trainers but they have a seriously low ROIs like around 0.5 does that not give a statistical advantage of betting that race? Ideally betting other horses in that race. Or all of the others with a super high rebate. If 50% of the pool is on trainers that have an ROI of 5 doesn't that give you a 25% advantage in a long run statistical situation. Just a thought kick it around kick in the head

[classhandicapper]

I would double check their ROI with short priced FTS. It might be different than overall.

[thaskalos]

Are you still in the vape business Teddy?

[davew]

maybe, but dutching will be difficult. I have noticed that during the last minute and running of race, the eventual winners odds drop near 70% of the time.

[Gorrex]

Would you guys be interested in a feature to dutch wager on projected odds?

(TRU Odds on our platform)

[green80]

Originally Posted by davew

maybe, but dutching will be difficult. I have noticed that during the last minute and running of race, the eventual winners odds drop near 70% of the time.

While that may be true in some cases, I doubt that that is accurate overall.

[castaway01]

Originally Posted by green80

While that may be true in some cases, I doubt that that is accurate overall.

Why would that be hard to believe, that odds on the eventual winner drop 70% of the time and rise 30% of the time?

[Dave Schwartz]

Originally Posted by green80

While that may be true in some cases, I doubt that that is accurate overall.

Actually, the last time I captured some figures using gate odds to final, they were:
(This was 2018, I believe.)

74% bet down
26% same or bet up

Definition of down/pup based upon 1 tick of odds movement.

I have on my agenda to do it again but for me it is a manual endeavor. Even a couple of hundred races means watching... well, a couple of hundred races. LOL

The higher the rebate available at a given track, the more volatile the movement. (... because the higher the rebate, the greater the percent of pool that is wagered by the whales.)

[MJC922]

Originally Posted by Gorrex

Would you guys be interested in a feature to dutch wager on projected odds?

(TRU Odds on our platform)

More options are great, me personally I'd like a way to make a conditional two horse parlay based upon projected odds off the DD of the previous race. TRU odds is close to the the same if you allow conditional on TRU Odds at like 15MTP or 20MTP, that's probably close enough to DD for me to work with it, my understanding is DD comprises most of TRU anyway that early in the wagering cycle. Prevents me from having to be online to process the matrix from the prev race and allowing the ability to parlay would take me out of being forced into betting doubles (consecutive races) I can then just pick two good races on the card even if they aren't consecutive. Anyway, thanks for asking... my two cents. Dutch? Maybe. Two race dutch parlay? Now it's getting interesting but maybe difficult to program on the ADW-side I'm sure. Food for thought though.

[Gorrex]

You can already use TRU odds as a condition for all wagers.

In the pool selection just change it to TRU while building the condition.

[MJC922]

Originally Posted by Gorrex

You can already use TRU odds as a condition for all wagers.

In the pool selection just change it to TRU while building the condition.

As you might expect from my posts out here I would strongly prefer first half of the DD probables matrix projected odds as the condition. I don't quite understand the reason behind offering TRU odds but not DD odds standalone. TRU odds may be a bit more vulnerable to volatility and / or pool manipulation than DD standalone IMO as the DD is a finalized pool. We already know the win pool is volatile, if I merge that with finalized DD odds I'm not sure what it buys me, I doubt it's additional accuracy. I guess TRU odds offer something rather than nothing if a race isn't the second half of a DD. That's the only reason I can think of to offer it.

[Gorrex]

Yes the live pools are largely added in to give data on races where no past DD exists.

When the past DD exists its weighted 3x over live so unless there is significant outliers its impact carries pretty well.

In the past I added in some willpay data but honestly the amount data lost in willpays just didnt make it worth much very often.

Someday maybe I'll get PK3 final matrices.. if I do they'll be added in the same as DD but for both races..

[teddy]

I don't have my retail anymore it's all wholesale.

[teddy]

Well we got off topic there but I don't have enough data cuz I cleaned my computer to really look at races where you have two trainers with really low ROIs that are on the favorites in the race. That would be interesting low ROI of two horses in the race and then you take the rest of the horses on your wager and some kind of Dutch fashion. Just saying that hey there's a high percentage of this money that based on history is down the drain

[MJC922]

Originally Posted by teddy

Well we got off topic there but I don't have enough data cuz I cleaned my computer to really look at races where you have two trainers with really low ROIs that are on the favorites in the race. That would be interesting low ROI of two horses in the race and then you take the rest of the horses on your wager and some kind of Dutch fashion. Just saying that hey there's a high percentage of this money that based on history is down the drain

I apologize Teddy for the thread veering off topic. It's actually some important info from Gorrex that hopefully others have picked up on as they can leverage much of the power of the DD projected odds using TRU odds without doing any of the fancy spreadsheet number crunching I do as part of my routine. As for trainers I'm not much of believer in ROIs when it comes to trainer stats. A lot of low ROI stats are just as much anomalies as anything negative IMO, you'll often see a low ROI with few wins and then a boatload of places, a hundred races later things may just as easily swing in the other direction more wins than places. Ymmv. The good thing is it should be a 'testable' angle if anyone has the DB for it.