Runs of losses.

Hi,

Is it possible to statistically calculate the amount of losses you have when you have a know strike rate?

ie if a system has an accuracy of 65% and a proposed take back on the losing stake is proposed?

eg: if my stake was 1.00 and the next race I propose to cover that losing stake by upping the new bet to include the loss from the last race?

What is the probability of having 1 - 20 losses in a row?

Please Help!

Hi PlacePlayer, and welcome to the site!

I could certainly answer your question with mathematics, if you could provide more information (which in regards to 'races' might be difficult for you to do). The probability of losing X in a row in ANY event is simply that probability to the events power. Let me give you an example:

In blackjack, the player has a defined chance to lose of 49%, win 42%, and tie 9%. So... what's the odds the player loses 5 hands in a row? Well that's the odds of losing (49%) to the power of the number of times (5)..

Probability of losing 5 in a row in blackjack = .49^5 = .02825, or about 1 in 35.

So now you're asking what's the odds of losing X races in a row... Well, in order to define that we need to know what the defined % probability of losing 1 race is. With racing, I would think it's a bit too dynamic to calculate, unless you run all the odds and trust the odds makers are accurate.

For example: Say you review the odds and decide the top 1 favorite of a race will win 50% of the time (I don't know racing that well just bare with me). Well, say you bet on that top favorite to win. The odds of you losing are the following:

1 Race: .5^1 = .5 (50%) of the time, 1 in 2.
5 Races: .5^5 = .0313, or about 1 in 32.
10 Races: .5^10...
20 Races: .5^20... and so on.

What you need, which again I think you might not be able to get, is your actual odds of winning/losing with your bet on EACH race. Above this assumes the EXACT SAME dogs/horses/etc run each race... Normally in the next race there's all new dogs/horses with all new odds! I hope I shed some light on how the math works, though I believe you'll have a tough time calculating this for very specific races.

It can still be done, but it would involve a lot more information (looking at the odds of whatever you bet on each specific race) and then multiplying all of them together over the number of races you want to know how many you could lose in a row...

Probability of losing race 1 = X
Probability of losing race 2 = Y

Probability of losing race 1 AND race 2 = X * Y

(and so on)

Hi Romes,

thanks for the great response.

The average odds on all these instances is 1.30 and the actual strike rate is 66% of winning a place.

Thanks!

Well then... if you know for EVERY RACE you have a 66% chance of winning and 34% chance of losing, then that's something to work with...

1 Race = .34^1 = .34...

5 Races = .34^5 = .0045...

10 Races = .34^10 = .000021...

etc.

Very Interesting stuff.

Would it be possible to calculate the likeliness if I was to have 15000 bets, what are the chances of having x amount of losses in a row?

something like this?

Z

I'm 1 for 34 on flush draws today if that means anything.