Bitplayer asked a great question. I don't claim to have the complete answer but I do have some opinions.

My response really has 2 parts. The first part is a comment on making odds lines and the other is a comment on how to use them to make betting decisions. None of this is particularly original and if you are really interested in this topic you might want to read one of the books by Dick Mitchell or Marc Cramer.

First – I don't think many people (myself included) can make an odds line that is accurate enough to spot the difference between a 33/1 shot and a 50/1. The difference between them is about a 3% chance of winning vs. a 2% chance of winning. In fact, once you get to less than a 5-10% chance of winning, you probably should pretty much clump the rest of the field together. Spruce Run is a good example, he looks like a big overlay but I really would not have much confidence that it is a true overlay or by how much. However, if the crowd bet him down to 10/1, I would be confident that he was a big underlay so it is not a complete waste of time to try and estimate the odds even for the longshots. I don't mind tossing in these types underneath in exotics (particularly third or fourth) but my rule of thumb is to avoid playing any horse to win that does not have at least a 10-15% chance to win.

Which brings us to the second part of my answer that deals directly with the question echoed by mholbert – how do you decide between playing a live longshot and a solid favorite?

To help build some intuition for the answer I will give a hypothetical example. Let's say someone offered you the following wager: he said the he would pick one card from a deck (with replacement after each draw) and he would pay off as long as that one card was not the Ace of diamonds and he would pay off such that you had a 10% edge on every bet. In other words, you would cash 51 out of every 52 draws (on average) and you would make a long-term return of 10%. Obviously, you would take as much of that action as you could get and you would very likely start off making a profit on the very first bet and steadily increase your profits as you play with only an occasional loss.

Now let's pretend that the same person makes an offer to change the rules of the game and says that in the new game he pays ONLY IF the card drawn IS the Ace of diamonds. He still pays such that you have a 10% edge but now you only cash 1 out of every 52 draws. Both games have the same edge and thus the same long-term expected ROI but which would you rather play? In the second game you could easily start out with a long losing streak before you ever cashed. In fact, unless you had a big starting bankroll, you could easily tap out before ever cashing a single bet.

The basic principle here is that you should factor in both the expected return (edge) and probability of cashing (reflected in the odds) when making a wagering decision. The higher the edge the better and the higher the probability of cashing the better. So a 2/1 shot that you think provides a 50% edge is a much better bet than a 50/1 shot that provides the same edge. But how much better?

There is actually a simple formula referred to as the "Kelly" formula that answers that question. The formula is EDGE/ODDS; the higher the better. For example, the 2/1 shot with a 50% edge has a Kelly value of 0.50/2 = 0.25 while the 50/1 shot with the 50% edge has a Kelly value of 0.50/50 = 0.01. The 2/1 shot is a much better bet. In fact 25 times better!

That's a long answer to the Bitplayer's question. The shorter answer is that I am not even very sure that Spruce Run was an overlay and, even if he was, he was not a very good win bet.

Hope that helps and again thanks for all the nice comments. Doing the ROTW is not easy even when you have a strong opinion in a race and it is even tougher when you don’t really have much of an opinion. I am just happy that I did not completely embarrass myself. I want to thank Mall, Catalin and OPM as well as the TG staff for their help in my first foray into the ROTW.

Chris