• La Coronel (5-1) leads them all the way in the Grade 1 Queen Elizabeth II Challenge Cup.Posted 4 days ago
  • Rubilinda (6-5) finds the wire just in time to take the Pebbles Stakes.Posted 4 days ago
  • Engage (1-2) rolls home from last to win the Grade 3 Futurity Stakes.Posted 4 days ago
  • Bolt d'Oro is the 12-1 favorite in the current Las Vegas line for Kentucky Derby 2018.Posted 7 days ago
  • Romantic Vision (6-1) takes the sloppy Spinster (G1) at Keeneland.Posted 10 days ago
  • Unique Bella (1-5) returns with a clear victory in the L. A. Woman (G3) at Santa Anita.Posted 10 days ago
  • Flameaway (5-1) wins a three-horse photo finish in the Dixiana Bourbon (G3) at a wet Keeneland.Posted 10 days ago
  • War Flag (9-1) wins the stretch battle in Belmont's Flower Bowl (G1).Posted 10 days ago
  • Separationofpowers (9-5) impresses in the Frizette at Belmont Park.Posted 10 days ago
  • Roy H (4-5) lives up to favoritism in the Grade 1 Santa Anita Sprint Championship.Posted 11 days ago
Breeders' Cup 2017
HRN Original Blog:
The Dark Horse

It's All In Black and White (and Red) – Part II

This is the second part in my series regarding calculating ROI. If you remember the first part of the series (CLICK HERE), I discussed an in-depth analysis of three years’ worth of races to see if I was actually making money and in which categories. In that analysis I learned that only 9% of my impulse bets turned a profit, that impulse (non-handicapped) bets cost me a lot of money over the years, and that I suck at straight wagers and perform much better with exotics. That analysis also provided a look at how I performed at certain tracks and with certain types of bets at certain tracks.

Well as my wife would say - I never do anything half-assed, so I decided to take a deeper dive with the stats. I poured through chart after chart, hour after hour, to see how my wagers performed in many other categories such as surface, distance, class, expanded race conditions, and even some goofy categories as well.

Here we go with round 2.

Now remember, in explaining the charts a little, there are three columns for each slice of data. BETS WON represent just that. The percentage of those kinds of wagers that actually won. BETS w/ PROFIT are bets that actually turned a profit. For example, if I wagered $5 on an exacta and it only paid $4.20, then that would represent a bet won, yet it was not profitable. The return on investment (ROI) is represented as a percentage. -100% would be a total loss, 0% would be break-even, and +100% would be doubling my money.

Also in this analysis I did NOT include my impulse wagers, tips, or going after bridgejumpers. These are only handicapped wagers (which will obviously make my numbers look better!) I did also include the horizontals with every analysis so that all wagers are reflected. Horizontal wagers span many different classes, surfaces, and distances. I haven’t found a good way to break those up and represent ROI by the different categories.

First I wanted to see how my ROI stacked up against the different race conditions possible.

No surprise to me here. I have always struggled with maidens and claimers and the numbers don’t lie. The ROI on stakes-quality horses is astounding at a positive 65%. For the maidens, it just doesn’t matter if they are maiden-allowance or maiden-claimers, the negative ROI leads me to believe that I should just stay away from those races entirely. In three-plus years I still haven’t been able to figure them out I guess.

Now out of curiosity, I wanted to know how I performed within the conditions of claimers, allowance, and allowance optional claiming. Here for allowance conditions...


For reasons I cannot explain, I do well with the N1X and N3X conditions, but NOT with the N2X conditions at the allowance level. I’ll need some help interpreting that.

And as far as claiming conditions…


Well other than some real goofy conditions, or combined conditions (C), my ROI is still bad regardless of condition. So then, does the claiming price matter then?




Again no surprise that when the claiming price, for both claiming and allowance optional claiming, is higher, my ROI is better. Oddly enough, I broke even with the cheap $5,000 claimers. I would think then that the higher the purse of the race, the better my ROI would be, right?


As I suspected, the “richer” the race, the better my ROI is. Should I just stay away from the cheaper races all together? Or perhaps lump them in with the horizontal wagers?

Next, I really wanted to see how I fared on different surfaces. Initially my guess was that I was good with dirt, mediocre on turf, and awful on polytrack. Well, my turf numbers were disappointing.


As I suspected, I do well on dirt, and I think I always have. I also suspected that I didn’t fare well on Polytrack, and the numbers confirmed that as well. However on grass, I have a disappointing -14% ROI. Man, I always thought I was better than that.  Then again, I'm RIGHT on turf 24% of the time, but is the object of the game to be right or to make money? Anyway, I wondered how my ROI changed based on surface conditions.


Apparently I like off-tracks. On the dirt seems like the wetter the better. On grass, I seem to prefer a wetter course as well. My ROI on fast dirt is just below break-even but I have work to do on firm turf.

So we’ve looked at class and surface now. What about distance?


I found this to be very interesting. The numbers show that I am much better at routes than I am at sprints. I would love to know WHY that is, but still interesting to look at. So of course in dark horse fashion, I broke the distances down to the specific distance which the races were run.


Give me the classic 10 furlong distance! I am almost break-even at 6 ½ furlongs and one mile. I don’t see much correlation there however.  So then the next logical step is to look at the surfaces and distances together. The results are pretty much what you would expect.


Routes on dirt is where it’s at for me, and I’m not that much under water with sprints on Polytrack. With all of this, I then wanted to look at how I fared with specific surfaces at specific tracks. I was thinking that maybe I’m good with turf at certain tracks or good with Polytrack at others. Here’s that slice.




































So hey! I do have a positive ROI on some turf courses after all! Looks like the grass at Hollywood (darn), Fair Grounds, Keeneland, and Meydan are pretty decent places for me to make some money. I’m also not as bad, in total, at Arlington as I originally thought I was either.

Wow, that was a lot of great data! Now you wouldn’t expect me to stop there would you? I also performed some analysis with some sillier categories as well. Nothing much to take away from these analyses, but still fun to look at.

I wanted to see if I had a positive ROI with any specific state-bred races.


So all that time in and around Chicago definitely gave me an advantage with the Illinois-bred horses. And wow! Look at those Florida and New Jersey-bred races! Abysmal is the word I would use to describe state-bred races at Woodbine and Fair Grounds. Yikes. Anyway, I will have to ask HRN's very own Matt Shifman about those NJ-bred races!

And here’s another weird slice of data. What’s my ROI based on day of the week? (I mean really, who would actually look at that data or even have the desire to calculate it?)


Well I should definitely be playing the races on Fridays and Saturdays! And as far as Mondays and Wednesdays? Sheesh, I won’t quit my day job!

How about a look of ROI by numbers of starters in a race? Like that would mean anything?


I guess I like races with many options! Or not so much with the 20% ROI in races with just five starters. And with just five starters, you would think that ROI would be better, right? As a lifetime horseplayer I should be able to pick a winner with only five options, right?

And here’s another odd look at ROI based on race number…


Haha! I should just bet the first race of the day (on a Friday) and leave the rest alone! Actually for the first four races of any card I seem to do pretty well. Then what, do I just get bored of handicapping races? Weird.

Anyway, with all of this, it does appear that my handicapping has improved over time. 2013 has been a great year so far and I hope that it continues through to the Breeders’ Cup!


What did you think of this analysis? Have you ever looked at your wagers in this manner?
Remember, this is part II of the series.  Click HERE for Part I.  

I do pose this question to all of you however...is the object of the game to be right, or to win money?  Discuss...



comments powered by Disqus

Older Comments about It's All In Black and White (and Red) – Part II...

Outstanding. Way to go my man.
Wow thanks!! I'm joining Twin Spires' own Derek Simon on his podcast tomorrow to discuss in more detail!
Yes I did. I read it when you first posted it, but didn't put two and two together until I saw the spreadsheet layout that you had made. Awesome job...I'm definitely impressed.
Thanks Jay S! Yeah I am telling the whole world here that I lose way more than I win! However I now know where I win and that is important. It's also good to know where I was not winning in cases where in my mind I thought I was. I think I'll go back and do odds in part 3... did you see part 1 too?
cpsoren, this is an amazing compilation of data. I admire the dedication and effort it would take to log this much information. It's interesting to see that you, like me, are better at handicapping dirt than turf. When I was going to the races regularly at LaD, I got to a point where I would almost never wager them. To me, turf races are filled with "blanket finishes" and it's difficult to get-up for the win. The only additional item that, to me, may be value added is a tracking of the odds played. Also, a tracking of the difference between Morning Line odds and post time odds. cp, this is an astounding endeavor you've undertaken. Good Show! Oh, BTW...I also admire the honesty of your presented analysis.
cps, my wife definitely wouldn't, hence the need for me to "re-take" statistics. :) Man, I was just about to shut-er-down when I saw your comment. I promise to take a look tomorrow, and give any feedback I may have. But, I promise, I'm pretty good at stats, but no expert. I do, though, normally agree with what tv says about probabilities and other statistical analysis.
Jay S. would you or your wife have any feedback on this blog post?
The truth is fleeting, it is always in flux. TO understand life one is constatnly learing after the fact...To sit still and accept the dogma is not to stand still but to go backwards. Had a old guy come up to a group of us with laptops at the track and he started out with that TIRED old statement "Well I Bin comin to the track here for 40 years and I never-------------------------(fill in the blank)" No, we said, you came to the track ONE year and simply repeated it 39 times!!
My better-half is working on a second degree and is currently taking statistics on-line. What does that mean? Means, once again, I'm taking statistics. Spent 10 solid hours Saturday refreshing myself with probabilities, variance, standard deviation, Poisson (sp?) formula, etc.
Many's the lass that wished you would just give it up.
Many's the lad that hates a proper analysis of statistical nebulism.
Chris, nice work. Travel Vic, thanks for the chuckles on a bit of a slow day, now it's time to SHUD-UP!
Chris, fantastic analysis! Sports bettors do this, so why not horse players. Some gamblers only bet certain conferences, because they have a better understanding of the teams. It makes perfect sense for players to focus on whatever they're strength is turf, dirt, claimers, stakes, etc.
MORE people need to understand the great investment gem HEDGING.....http://www.drf.com/news/steven-crist-some-fourstar-hedging-advice-15k-winner
I have hit rate and profit rate AND ROI for EACH category in the last 2 blogs.
EACH race track in North America has an energy distribution finger print...Horse does not fit the standards, the harder it is to win....REAL basis for the old horses for courses
I got the impression that these data were desinged all around $net, which is the end product of the game after all
Travel_Vic, I think you're getting way too deep in the minutia of statistical analysis. First of all, you say that one needs 10,000 races/horses to establish the significance of a factor and, then, you quote Quirin, whose studies typically encompassed far fewer races/horses. What’s more, impact values and pool impact values are designed to measure handicapping FACTORS and are not really suited for the kind of broad analysis that Chris has presented here. Your desire that he “break down” his data can actually be counterproductive to Chris’ future betting endeavors. In the absence of VERY LARGE sample sizes, breaking down historical data often leads to the “backtesting fallacy.” Remember, Chris gathered his data in a completely random and unbiased way. By searching for niche betting opportunities in the future — identified by breaking down the (insufficient) data he has — his future data will no longer be random and unbiased. And, in my opinion, he will probably not substantially improve his game by doing this. I am always suspicious of handicappers that say they dominate at Track A, but suck at Track B, because, generally, they are basing that opinion on very small sample sizes that can change on a dime. Is Mountaineer different than Saratoga? Of course. But the difference is largely due to the quality of racing, not geography as you imply.
you can have a hit rate of 40% but if the average mutuel is $3.60 you will lose long term..Hit rate o 25% at 12.60 is sustaining over time
% success is vague

Connect With

"The Dark Horse"


Me On Facebook
Follow Me On Twitter


 (Whoa - it's me - the DARK HORSE)
Greetings and welcome to my "dark" blog!  I have been a fan of horse racing since I was about 12 years old.  Growing up in Minnesota, my dad would take me and my brothers to Canterbury Downs and I would practically beg to go every weekend!!  I made my first "big" score there hitting the win, place, and exacta in the same race at age 14 (pulled in $84 with horses named Railroad Red and Racers Dream!).  I even had a poster in my bedroom of Lost Code winning the 1987 St. Paul Derby!
While finishing school in Chicago, I spent every possible moment at Arlington Park with my dad.  We took the time to develop computer programs (on the ever popular Lotus 1-2-3 for DOS) to try to find patterns and handicap races more accurately.  My dad taught me a lot about handicapping races, paying particular attention to class moves!

Since then, I have become more involved with horse racing each passing year through a variety of different media-based roles. Once deemed the "second best racing ambassador in Chicago" (hey I've been called worse), 
I now reside in the Mid-Atlantic Region of the USA and will be covering many different aspects of racing - in my unpredictable DARK HORSE STYLE!  
My favorite tracks are:
Del Mar
Kentucky Downs

Some of my all-time favorite horses include: 
Feel free to contact me anytime and I will respond! 
Twitter: @GoDarkHorse <---Preferred
Here are some fun things to check out! 
Saturday Morning Radio Program: Click Here
My Main Web Page: Click Here
Speaking on Derek Simon's Show: Click Here
First time hosting radio - Hour 1: Click Here
First time hosting radio - Hour 2: Click Here
Second time hosting - Hour 1: Click Here
Second time hosting - Hour 2: Click Here
With Derek Simon again: Click Here

Here are some of my racing photo albums!

And here are all of the tracks I have visited in person!

This blog is eternally dedicated to the memory of Tim "Tiznow" Reynolds.

Top Stories