Sunday, December 6, 2009

Saturday is a dog day on ice, with 10 wins in 14 hockey games. Blind luck, or another sign that the math always corrects itself?

This just in: "You can't take one good day, or even one bad one, and claim that it proves anything. What will happen to this daffy dog idea of yours is that over time, lucky days that put you ahead will be outnumbered by regular losing days, and you will end up so far in the hole that you will never be able to get out of it."

Let's first of all look at Saturday, December 5, which saw a 70% DWR on the NHL schedule and a more "regular" 44% (4/9) in the NBA lineup.

The total win in the 7-dog trial was $2,650 - not quite enough to get us out of the hole, but it left us hanging onto the edge of it by our fingertips, ready for the boost we need to pull us back into the black (pardon the mixed metaphor).

I have been saying for weeks that slumps in the overall DWR have to be rectified at some point because the bookies are counting on it, and those folk don't count on anything that isn't a copper-bottomed certainty.

When underdogs get beaten up, so do the bookies, and no one making book is doing it to put extra money in the wallets of punters who like to play it safe by backing favorites.

Since this test began on November 1, there have been 16 losing days and 19 winning ones, with the DWR exceeding 40% only 12 times in 35 sessions.

When target betting was first developed for casino table games, which have a generally higher win rate than underdog teams in American ballgames but very rarely pay more than even money on a bet, the whole objective was to recover losses in far fewer bets than it took to lose the dough in the first place.

That remains the paramount goal when the method is adapted for the sports book.

Yesterday, Saturday, we placed our bets hoping for relief from a slump in which we lost more than $4,000 after coming within a couple of hundred bucks of setting a new all-time high.

The damage exceeded $1,000 a day, slightly relieved that week by two days in which we trod water, winning a total of $650.

Target betting assumes that we are always going to lose more bets than we win, so bet values are geared to the need for a quick turnaround when the time is right.

It is quite possible that bets for today, Sunday, will all be flushed down the toilet, setting us back $2,800. But it's not likely simply because, overall, underdogs have maintained a win rate of around 40%, delivering an average of three wins a day.

Because "dogs" pay at least even money and usually better, each win recovers enough to offset one loss, sometimes with change to spare.

Three wins at average odds of +135 will offset four losses, making it a break-even day for the 7-dog trial, and four wins will push us up the ladder a rung or two.

There is nothing unusual (or "lucky") about days with five or six dog wins out of seven games, especially after several days in which underdogs underperformed!

The charts below should help add a little perspective to this never-ending argument in which dognostics keep reciting their mantra that losses are the norm and wins are anomalies.

The first summary tracks games from November 1 in which all dogs within the +100 to +180 range are bet, but the target betting rules (meaning bet values) are applied as if the games were consecutive.

Statistically, there is nothing wrong with treating the games that way, even though it would not be possible in reality.

We are simply treating the outcomes as samples from reality that are in essence random and objective, since we do not determine which team wins or the odds applied.

The second chart tracks the same outcomes in however many separate lines or series would have been needed to enable us to expand the 7-dog trial to cover all bets falling between +100 and +180 on any given day.

As always, these charts and summaries don't prove anything. They simply show what can happen, and probably (but not certainly) will.

Click on an image to enlarge it

An important reminder: The only person likely to make money out of this blog is you, Dear Reader. There's nothing to buy, ever, and your soul is safe (from me, at least). Test my ideas and use them or don't. It's up to you.