I want to stress this is not really an interview.
Dave: My very special guest today is Derek Simon. We actually started this conversation about an hour ago and I wished I had just turned the recorder on from the very beginning. I know you all know Derek so this is going to be a great conversation.
We are just winging this but we had this tremendous conversation before I started the recorder so we are going to try to recreate it the best that we can. It started with Derek talking about spot plays. So Derek why not just take over from here and I will interrupt you as I usually do.
Derek: For me over the last 5, 6 years I haveÂ really gone in the direction of database handicapping, accumulating
as much data as I can and letting that guide me.
For some reason I just do not like to handicap much anymore, and I used to live that. You wake up, have the coffee and the Racing Form and now with the family I just do not have as much time so I love doing the database.
Spot plays to me have been a really interesting journey because there is so much that I think you learn about, what is a proper sample size and how you go about actually parsing the data, I do that wrong all the time.
We are talking about people with angles and they have 100 plays and it started with 200 and they get all these other factors that they throw in there. It is not enough, those things invariably do not hold up. I had to learn that all along the way in this journey because those sample sizes from Quirinâ€™s book were not big enough to draw some of the conclusions that he did. It has been a very interesting journey and I love that approach to the game.
Dave: You are right. In Bill Quirinâ€™s book, all of his databases combined (as I recall) was around 2,200 races and that was gigantic for the time. What do you keep your database in?
Derek: It is through Excel, and I have automated it as much as possible. I am always very envious of you. I envy the fact that you have so many races.
Dave: I have about 450,000 races currently. But 50,000 races is really plenty. Weâ€™ve spent a little time talking about sample size. When you are building systems to prove something really works it probably takes 2,000 – 2,500 bets.
But it also depends on the odds range we are talking about. If we are talking about a sample being carried by high odds plays you have to have that 2,000 to 3,000 races. If you are focusing on medium to small odds plays, you can use a substantially smaller database. You use Excel and you have 50,000 races right?
Dave: Do you use the filtering to sort on these factor columns you have?
Derek: Basically I start with just the data. I have always been a proponent of get as much as you can and you can always decide what you do not need. Where that causes a problem, I have learned that Excel is incredibly temperamental once you get a lot of data.
One of my projects recently was to consolidate all those databases to the stuff I was primarily looking at because there were those in there that I would never need. Things like weight for the last 10 races – all kinds of different things – that I really, at least at this point, have not spent a lot of time studying.
I recently condensed everything and then I will just run queries and look for certain factors, of course I have a lot of my own proprietary and just kind of do my testing and back testing that way
Dave: How has this worked for you, what kind of hit rate are you seeing, what kind of ROI?
Derek: It is interesting because you talked about low odds and ROI gets such a horrible bad rap. I constantly hear people talking about value. They say that you make your real money on horses that are 12-1 or greater.
Thatâ€™s okay, but as you and I both know you start doing that and what happens? Your hit range necessarily goes down. There are people out there that I seriously think believe they can maintain a 1000% or 100% ROI. It is not possible, I do not strive for that.
What I do strive for is consistency and I think the best way to achieve consistency is in those lower odds horses.
I generally like my angles to have a 40% strike rate or greater. There are a couple of things that cause inconsistency in horse
racing. One of them is your hit rate, the other â€“ which also defines your Kelly advantage – is your odds.
As you raise that your percentage you also lower the odds, which means greater consistency.
About Derek Simon:
I love numbers and analysis and have been able to parlay that into an exciting career writing about two of my greatest passions â€” business and sports.
In addition to working as a freelance financial writer for Newsmax, The Motley Fool, Investopedia/Forbes, Beacon Equity Research and Investor Concepts (among others), I was also the editor of Small Cap Insider, a monthly newsletter highlighting investment opportunities in the small cap sector.
Currently, I am the editorial director for US Racing where I contribute written, audio and video content and oversee a team of talented and passionate writers.