Market losers

When you’re considering doubling, one thing you need to keep in mind is if it’s necessary to double at this moment. Typically, you want to delay offering the cube for as long as possible, to creep as close as possible to Point ③ in the spectrum at the very right edge of the doubling window. This maximizes your equity while minimizing your risk, as you would be minimizing your opponent’s chance at a comeback.

However, if you wait too long, you might cross the threshold and enter into D/P territory. If this happens, you lose your market. You would much rather have played on for 2 points, but now are only winning 1 instead, which is an inefficient use of your lead in a game.

The cube decision for many positions can be described in terms of searching for market-losing sequences. O’Hagan’s law states that you should double if at least 1/4 of the sequences (that is, your rolls, followed by the opponent’s responses) are market losers. The position from the last lesson is a perfect example of doubling before losing our market:

17 of 36 rolls will hit Gary’s checker on the 7-pt, guaranteeing us the win. If we don’t double before this roll, we will have lost our market by a long shot if we do hit! 17 is larger than 36/4 ~ 9, so you have a strong double.

In other situations, having no market losers is a clear indicator that you should hold off on doubling. This concept is easier to digest with an example: suppose you know for a fact that you’re leading, but you also know that you aren’t quite at 75% win chances yet, either. If, when looking ahead, you see that the odds are that nothing will substantially change in the next roll (i.e., no market-losing sequences), you might as well hold off on doubling. By waiting, you are protecting yourself against the unlikely sequence where your opponent rolls a lucky double and gets out of their situation. If, however, things proceed at their natural course and improve gradually, you can then double at a stronger situation. The only downside to waiting, really, is the possibility of losing your market, so if you see no market-losing sequences, you shouldn’t double!

Next lesson: Recube vig