Betting the favorite to Place

[Ray2000]

Betting the favorite to Place has the best ROI percent (-9.0%) in this group of ~5200 races Feb-Aug '14
Now if there was only some way to toss out the bad favorites....


Does any ADW offer conditional bet on Fave if ...(criteria)



also:
In a small sample of 27 races, only 2 showed a change in the favorite from 0 mtp to closing odds.

[mrroyboy]

Well that's why I have tried to start discussions etc. If we can come up with elimination guidelines etc and throw out those bad favs we turn that into + roi. Good for exactas and more.

[imofe]

I think it can be done. But it is not just separating certain favorites from each other. I believe it is more situational. Not just the ability of the horse. I would look for horses that will be on the lead at front favoring tracks and are most likely not going to be pressured by an all or nothing brush type horse. We also don't want our horse to be a win or fade horse. Obviously, trotters are much more likely to break. This may not be a big deal if you are making a win bet on a 5/1 trotter. But if you want win/place consistency, avoid these races. This is just scratching the surface on what can be a very detailed project.

[Sea Biscuit]

Originally Posted by Ray2000

Betting the favorite to Place has the best ROI percent (-9.0%)

Ray: Just one question.

How can it be good if the said ROI is in the negative.

[Ray2000]

Sea, not "good", just best of the group. -9% ROI is almost break even with rebates...

imofe, good point ...avoid the "Silky Sullivan" effect.

roy, that's why I was looking at this data.

There's a recent thread* in the T-bred section where 'shouldacoulda' states
"...The trick is spotting one.."....how true

But false faves in Harness is a different game.

We have driver ability, breakers, potential traffic problems and early speed placement,
all which I believe to be more important in Standardbreds vs T-Bred because of the uniformity of the distance.


In my last "screwy" run** had 'PLACE' bets on robot picks outperforming the 'WIN' bets on ROI but it had nothing to do with crowd choice.

I'll start a new run here, with the same robot pick method but only betting "PLACE" if the bot pick is favorite.
(Chester race is over and and that one is redboarding but give me that one OK gang?. ..$3.80)


Aug 21, 2014
Hoosier R6............# 4 Always A Diamond-7/2
Hoosier R10...........# 5 Ks Hall-3/1
Hoosier R11...........# 4 E R Oliver-10/1
Mohawk R10............# 7 Little Red Chev-15/1
Northfield R7.........# 5 Naked At The Beach-4/1
Northfield R9.........# 3 Vodkainduceddreams-7/2
Northfield R12........# 4 Sturdy'S Baby-7/2
Chester R4............# 5 Wisenheimer-5/2
Saratoga R5...........# 6 Canamerica-9/2
Saratoga R8...........# 4 Brickyard Nellie-3/1
Yonkers R1............# 1 Putnams Glory-3/1



**
http://www.paceadvantage.com/forum/s...=115236&page=4


*http://www.paceadvantage.com/forum/s...d.php?t=113961

[Poindexter]

2 things. What would happen if you eliminated bad posts from your sample? Anything outside of post 6. I would have to believe your roi would improve. Regarding getting rid of bad favorites, I doubt you have this information in your database, but I would start with any horse coming out of a race where he had a perfect trip(examples) had lead on a slow paced race, sat a 2 hole trip behind somebody that went too fast(especially the 3rd quarter-where the leader ends up burning out the first over horse and kills the overland train) or the other phenomenon a horse that goes first or 2nd over to a weak leader(maybe a longshot, or just a horse who doesn't fire) and the first over horse faces little resistance and the pocket sitters end up shuffled out of the race while the overland train has gained momentum. It is not always easy to get a trip so any horse who benefited from one in his last race would be the perfect toss in my opinion.

[imofe]

I think you make some good points about outside posts. Let's say you have a horse that just wired the field in a similar class from Post 3. Today he has drawn Post 7 and there is a big speed horse inside of him. A few bad things can happen that were not going to happen last week. He might never see the rail. He might be forced to tuck in a two hole behind a horse that might stop in front of him. But yet he is still the favorite and the most likely horse to win, but not a great place bet.

[Ray2000]

Poindexter, I checked inside faves (posts 1-6) but the numbers were about the same
In the data base races I used,
In 4455 of those races the favorite started from posts 1-6
1826 won (41%) returning $7577 (-15.0% ROI)
2747 finished 1st or 2nd (62%), returning $8073 for Place bets (-9.4% ROI)

imofe, You got that right
Trying to teach the robot on which horse(s) are likely to leave is really tricky.
I use the average position gained/lost from post to halfmile marker over last 6 starts,
adjusted for driver aggressiveness.




I'm still setting up this "Robot plays Chalk" run but yesterday's results were typical.
(I have to get away from manually checking Favorite status)
Naked At The Beach and Wisenheimer actually won but didn't go off the fave, so not counted.
I did use some Harrington and Monti chalk picks in setting up the spreadsheet.

..... Bet Return
Win 26 32
Place 26 29

Here we go for Friday, August 22, 2014
Scoring will be on $2 Place Betting IF the Pick goes off the favorite else pass.
Batavia R4............# 1 Good Luck Penny-5/2
Batavia R5............# 1 Helena'S Hope-5/2
Batavia R6............# 1 Bobby Be Good-3/1
Batavia R7............# 1 Milliondollardad-5/2
Batavia R10...........# 1 Soaring Honey-4/1
Maywood R1............# 4 Fox Valley Cupid-5/2
Maywood R2............# 3 Yanklet Hanover-8/5
Meadows R6............# 5 Myology-3/1
Meadows R14...........# 8 Global Magic-2/1
Mohawk R3.............# 3 Beach Gal-5/2
Mohawk R4.............# 5 Thatll Be Franny-7/5
Mohawk R9.............# 7 Lifetime Pursuit-6/5
Mohawk R11............# 6 Request For Parole-5/2
Northfield R1.........# 1 Laser Lucy-2/1
Northfield R4.........# 5 Umygeorgia Hanover-5/2
Northfield R5.........# 2 Classic Sakuri-4/1
Northfield R10........# 4 So Long My Love-7/2
Pocono R8.............# 1 Marathon Day-5/2
Chester R2............# 5 Gallie Bythe Beach-7/2
Chester R4............# 5 Bahama Blue-8/5
Chester R6............# 5 Shelliscape-4/1
Chester R8............# 2 Express Jet-8/5
Chester R9............# 2 Bondy N-9/5
Chester R12...........# 8 Also Encouraging-8/5
Chester R13...........# 6 Ya Want To Drag-8/5
Chester R14...........# 1 Jj Alex-8/5
Scioto R6.............# 6 Newbegining-3/1
Saratoga R3...........# 5 Rock This World-5/2
Saratoga R5...........# 4 Medicine Ed-5/2
Saratoga R6...........# 4 Pop I-5/2
Saratoga R8...........# 1 Morosita Bi-5/2
Tioga R6..............# 4 What About Brian-2/1
Tioga R7..............# 5 De Vins Girl-3/1
Vernon R3.............# 5 Morning Joe-5/2
Yonkers R12...........# 1 Virgin Mary-8/5

[imofe]

imofe, You got that right
Trying to teach the robot on which horse(s) are likely to leave is really tricky.
I use the average position gained/lost from post to halfmile marker over last 6 starts,
adjusted for driver aggressiveness.

Ray,
Can you elaborate on this. How many races do you go back? What spread range are you looking for that shows one is much more likely to leave? How is driver aggressiveness measured?

[mrroyboy]

Ray do you have any stats on these favorite's morning line odds? I am curious if "bet down" type favorites say 4-1 or better out perform morning line favorites.

[Ray2000]

roy
I have them but not sure how to present it, let me think about that.

imofe
When I said "It's tricky"... that was an understatement .
It's most likely the factor I use with the largest uncertainty.

What I do is use ALL 10 lines and get a propensity to leave by subtracting the call at the half from the modified starting post. (Modified means the 1st trailer is assigned post "1½" and the 2nd trailer is post "2½", etc).
This position change for each ppline is also modified by the driver's UDR rating and the driver's AG rating. (AG is my own numbers and is on a scale from 0 to Gingras ). I average these lines and call it ..ground gained/lost from start to half but it does ignore gaps. This is different from other calculations, where I filter out only the 'good' lines in the past 40 days.

If this value is large positive then the horse is leaving, large negative means a closer, in between is a stalker.
Another subroutine is then used to predict the effort the leavers will need to gain the lead.

[imofe]

Thanks Ray. It does sound tricky.

[Poindexter]

[QUOTE=Ray2000]Poindexter, I checked inside faves (posts 1-6) but the numbers were about the same
In the data base races I used,
In 4455 of those races the favorite started from posts 1-6
1826 won (41%) returning $7577 (-15.0% ROI)
2747 finished 1st or 2nd (62%), returning $8073 for Place bets (-9.4% ROI)


Very interesting. Thanks, for the info, I expected different. Is your database able to determine stuff like fastest last 1/4 in last race or 1 of 2 fastest last 1/4's in last race, or fastest final time in last race or fastest trackmaster class rating in last race(or maybe you do no even use trackmaster).....

I assume horse that were 7th or worse at the 1/4 pole have a dismal roi (obviously eliminate a horse that broke since he cannot win) or are you really going to shock me.

[Ray2000]

Poindexter
The data used for this analysis was derived from the charts, USTA and Standardbred Canada, and is not a good reflection of all NA races. It tends to be heavy weighted by Canadian tracks and some of the B tracks have pretty high take-outs on Win/Place bets. Then again, they break to the nickle which is significant when dealing with prices in the 2.20 - 3.00 range.

My interest was in the relative Win/Place roi's on faves so it suited me for that anyway. I needed to use the charts to find the 2nd and 3rd "Peoples Choice" horse, not available in a database of past performance lines.

As for your question on roi of closers I would have to use a different approach on that.




Yesterday... 35 Picks
26 went off the favorite, of those
17 won returning $61 .....roi-1.17
21 placed returning $53 ..roi-1.02

						
8/21/14 To 8/22/14 ---- 39 Favorites						
   	Bet	Return	Net	ROI	Hits	Strike
Win	78	89	11	1.14	27	69%
Place	78	82	4	1.05	33	85%

Still doing this by hand so might need corrections. Some race had co faves

Saturday, August 23, 2014
R. Aces R6............# 5 Black Hammer-5/1
R. Aces R7............# 1 Diva Las Vegas-2/1
Batavia R12...........# 1 Hudson Gal-7/2
Mohawk R5.............# 1 Slip Into Glide-5/2
Plainridge R3.........# 3 Happy Surprise-9/5
Saratoga R3...........# 1 Livefreeordiehard-5/2
Saratoga R5...........# 1 Aussie Reactor A-8/5
Saratoga R7...........# 5 Fameous Western-5/2
Saratoga R9...........# 4 Valentino-5/2
Saratoga R10..........# 5 Clooney Drummond-8/5
Tioga R2..............# 5 Lauxmont Capri-3/1
Tioga R8..............# 5 Bikini So Teeny-5/2
Yonkers R7............# 4 Rampage Jackson-5/2
Yonkers R10...........# 1 Doctor Butch-3/2
Yonkers R11...........# 1 Aslan-5/2

[Ray2000]

Yesterday 15 Picks
11 went off the favorite, of those
5 won returning $15 (-6)
8 placed returning $19 (-2)

						
8/21/14 To 8/23/14 ---- 50 Favorites						
   	Bet	Return	Net	ROI	Hits	Strike
Win	100	104	4	1.04	32	64%
Place	100	101	1	1.01	41	82%

Sunday, August 24, 2014
Pocono R8.............# 4 Dawson City-5/2
Pocono R10............# 7 Beat The Drum-5/2
Pocono R12............# 3 Liberal-5/2
Chester R1............# 6 Tiger'S Too Good-8/5
Chester R5............# 4 Explodent-4/1
Chester R9............# 1 P H Supercam-9/5
Chester R10...........# 1 Worth The Money As-2/1
Chester R11...........# 4 Chilli Nz-8/5
Chester R13...........# 4 Mr Hasani N-8/5
Plainridge R4.........# 4 Up Tempo-6/5
Plainridge R5.........# 1 Mordecai-7/2
Plainridge R8.........# 1 Fashion Shark-4/1
Tioga R12.............# 6 On The Big Swing-5/2