The Truth about Losing Streaks

It is, perhaps, one of the great mysteries of the world. No, I’m not talking about Donald Trump’s hair or Miley Cyrus’ belief that she is sexy. I’m talking about losing streaks in betting.

All of us except those who sell “locks of the week” (those guys are the best handicappers in the whole wide world) have experienced them: prolonged periods of loser after loser — like speed dating at a Star Trek convention.

I would like to say that handicappers can control these series of unfortunate events, but, sadly, racetrack gamblers in particular must simply learn to minimize the pain… or risk being consumed by it. 

Racing is not sports betting. In a typical point-spread play, a bettor has a 50 percent chance of winning and a 50 percent chance of losing, with the house keeping a 9.1 percent vigorish for facilitating the play. 

In pari-mutuel  wagering, there is almost always more than two betting options — except in off-the-turf events at Aqueduct and Parx — and the”vig”, or takeout, is in the 17-20 percent range on straight (win, place, show) wagers and in the 20-30 percent range on exotic bets. Hence, a bettor’s success rate and winning payoff are both negatively affected. 

What this means, of course, is a greater probability of long losing streaks — even if one’s overall winning percentage is relatively high.

Take, for example, my own spot plays in the latter half of November. From Nov. 16-30, I offered my best method and angle picks to those who purchased one of the report packages offered at my Web site and, all in all, it was a successful promotion. 

During the two weeks, I had 67 total plays, of which 48 met my fair odds (when stipulated). The results are below:

All Selections: 67

Winners: 30
Win Rate: 44.8%
Return: $151.10
ROI: +12.76%

Qualified Selections: 48

Winners: 20
Win Rate: 41.7%
Return: $115.60
ROI: +20.42%

Click HERE to see a day-by-day listing of all the selections.

Pretty good, huh? Well check out these “lowlights” and, remember, we’re talking about positive overall stats and a mere two-week period in which the plays were offered:

* One winner and a -83 percent ROI in my first 10 qualified picks (the selections were listed by track — in alphabetical, followed by race, order).

* After a great day on Nov. 23 (3-of-4, +127.5 percent ROI), I went 1-for-10, with a teeth-grinding five second-place finishes over the next six days.

So, in 14 days — 14 days! — I experienced two 1-for-10 skeins, despite a 41.7 percent winning average on the whole.

And you want to know something really amazing (if you don’t, please just play along)? These results practically mirrored another two-week promotion I did in September.

Take a look at the digits:

All Selections: 83

Winners: 37
Win Rate: 44.6%
Return: $183.20
ROI: +10.36%

Note: There was one off-the-turf race which did not count in the totals.

Qualified Selections: 56

Winners: 23
Win Rate: 41.1%
Return: $136.60
ROI: +21.96%

Click HERE to see a day-by-day listing of all the selections.

Once again, I present the “lowlights:”

* Starting with the 8^th race at Assiniboia Downs on Sept. 22 through Sept. 27, I cashed once in 18 bets (on contenders that met or exceeded my fair odds).

* Minus Saturday, Sept. 28, my win rate would have stood at 33 percent and my ROI would have checked in at a meager 1.67 percent.

The point that I’m attempting to make here is that, even with a relatively healthy hit rate (41.3 percent among the two promotions), losing streaks are the norm, rather than the exception.

This is true, perhaps to a lesser degree, in other sports as well. In the Denver Broncos first game of the season, Peyton Manning completed 27 of 42 pass attempts for 462 yards and an NFL record-tying seven touchdown passes. Yet, he started that game 1-for-4 and also had a 0-for-4 streak and a stretch where he was just 2-of-7… which brings me to part II of this treatise: the “gambler’s fallacy.”

The gambler’s fallacy is the belief among some bettors that their odds of winning go up with each successive loss. This is sometimes called the “Monte Carlo fallacy” based on an incident that occurred at the Monte Carlo Casino.

The story goes that on August 18, 1913, at a roulette wheel inside the famous casino, the color black came up a record twenty-six times in a row. According to reports, the Casino made millions of francs that day as players doubled and tripled their stake on red at various points during the run — spurred on by the fallacious notion that it was “due.”

To show the absurdity of this belief, I conducted a little test. I simulated 50,000 coin flips and recorded the results overall, as well as the results after five consecutive “tails” and 10 consecutive “tails.”

The outcome of my experiment was as follows:

TOTALS

Heads: 25,038 (50.1%)
Tails: 24,962 (49.9%)

AFTER 5 CONSECUTIVE “TAILS”

Heads: 797 (51.1%)
Tails: 763 (48.9%)

AFTER 10 CONSECUTIVE “TAILS”

Heads: 22 (48.9%)
Tails: 23 (51.1%)

As you can see, there is simply nothing to the idea that one’s winning chances go up — or down, for that matter — based on previous results. Once an incident rate or success rate (whatever one wishes to call it) has been established, each event is independent of any preceding or proceeding event. In other words, the flip of a (fair) coin is always a 50-50 proposition.

So what does all this teach us? Three things, I think:

1) Racetrack gamblers should focus on managing, rather than eliminating, losing streaks.

2) Each bet should be made independent of the last. In other words, no doubling up in the 8^th race because your “sure thing” in the 7^th lost its jockey at the gate. (Toward this end, it is wise to map out one’s wagers before the races.)

3) Don’t play/pass races arbitrarily; either you have an edge or you don’t. And if you don’t know your edge, it’s safe to assume you don’t have one.