Losing strategies wanted

[psycho040253]

Thanks Psycho

Here's another possible losing strategy. When the race goes in-play, offer to lay each horse substantially below its BSP.

My hunch is that the market will take the value and leave the rest. Also, let's say a horse drops suddenly from (say) 2.2 to 1.8 (perhaps because its rival has just fallen), and you had money waiting to be matched at 2.14. Someone would get to back at 2.14 when the true odds are much lower, at your expense.

I've just been number crunching a spreadsheet relating to ~1,400 horses (which includes min in-play odds), and found that blindly laying the sample at prices substantially lower than the BSP would have led to a hefty loss.

What do you guys think?

Jeff

Agree 100% that this strategy will lose. However, although I've never tried it, laying a single horse, whose price has crashed early in running, may give a decent return. Very often, the odds will rise sharply again, shortly after the fall, allowing one to trade out for a nice profit.

Psycho

[Iron]

Thanks Psycho

Possibly. But how do you translate a concept as vague as 'early in running' into a rigourous system?

Would you look at the typical winning times for (say) a Class 5 six furlong race, and say 'I will class anything up to x seconds into such races as being early in running'?

Also, what you're proposing is apparently predicated upon the notion that the market over-reacts to what it sees in the early stages of a race, attaching too much significance to early events. It's any interesting theory (and I appreciate that you're only presenting it as a theory), but I'd need to see some evidence to be convinced.

BTW, something I've noticed is that, if a favourite drops sharply in price within seconds of the off, it tends to do very well. That's not something I've studied scientifically - just a general impression.

Jeff

[psycho040253]

Thanks Psycho

Possibly. But how do you translate a concept as vague as 'early in running' into a rigourous system?

Would you look at the typical winning times for (say) a Class 5 six furlong race, and say 'I will class anything up to x seconds into such races as being early in running'?

I think that it would require some research to obtain a definitive answer. Either that, or just use one's gut feel and accept that the approach isn't terribly rigorous.

Also, what you're proposing is apparently predicated upon the notion that the market over-reacts to what it sees in the early stages of a race, attaching too much significance to early events. It's any interesting theory (and I appreciate that you're only presenting it as a theory), but I'd need to see some evidence to be convinced.

Yup.

BTW, something I've noticed is that, if a favourite drops sharply in price within seconds of the off, it tends to do very well. That's not something I've studied scientifically - just a general impression.

Yup - I've noticed that too.

The question is - does it do sufficiently well often enough to more than offset the losses incurred when it doesn't?

Psycho
I may not be in hell, but I can see it from here!

[tomallen123]

In running, put an order in to back every horse in the race at 1000.
What's my prize?

[Iron]

It's an interesting idea, but not necessarily a losing strategy, given that you get stuff like this happening from time to time: http://www.betangel.com/blog_wp/2012/01 ... t-taunton/

Jeff

In running, put an order in to back every horse in the race at 1000.
What's my prize?

[Ethanol]

Change the 1000 to 1.01, and you might well have a losing strategy.

[Iron]

Either that, or just use one's gut feel and accept that the approach isn't terribly rigorous.

Whilst there are members of this forum who make very good livings from their gut, I'm not a big fan of that approach.

How can you test something that isn't precisely defined, or be confident of replicating it?

Jeff

[psycho040253]

Either that, or just use one's gut feel and accept that the approach isn't terribly rigorous.

Whilst there are members of this forum who make very good livings from their gut, I'm not a big fan of that approach.

How can you test something that isn't precisely defined, or be confident of replicating it?

Jeff

I'm a big fan of the disciplined approach - a big fan.

However, I have an ever-growing small, special bank that I use for those 'gut feel' bets. It's for those 'I've got a funny feeling in my gut about this horse that I can't explain' times. The feeling comes when it comes. I wish I could replicate it at will. As for testing it ... I have an ever-increasing bank.

Over the years, I've learned that, if my brain says one thing and my gut says the opposite, if I don't go with my gut, I regret it.

One more thing: Learn the difference between gut instinct and emotion.

Psycho
I'm not in hell, but I can see it from here!

[switesh]

Ok, I've found the Utlimate Losing Strategy. Here's what I did today:

For the first 10 races:
Lay top 4 runners (according to favoritism) at Best Market Price
Fixed Stake $10
Green 5 secs before official off.
Nothing else.

10 loses in a row, how's that for a hat-trick?
-$16.63 down so far.

I think someone in the post mentioned that "Do the opposite of a losing strategy, and you will have a winning strategy". Well you were wrong my friend (mean it in a friendly way, no harm intended . But either-ways I decided to do the opposite knowing full well that the opposite of losing - can still be losing.

So here's what I did for races 11 and 14:
Back top 4 runners (according to favoritism) and instead of taking Best Market Price I decided to improve it by asking for 1 Tick Above Best Market Price.
Fixed Stake $10
Green 5 secs before official off.
Nothing else.

Now, the average size of loss began to get bigger. In just 4 races the new loss was -$11.45 already. Phew, this couldn't go on so I decided to change tactics.

So here's what I did for races 15 and 21:
Back the top 2 Favorites and Lay the next 2 (still trading 4 runners overall)
Fixed Stake $10
Green 5 secs before official off.
Nothing else.

Ok, had 2 fluke wins in races 19 and 20 but the overall size of loss was still large for the ones that lost.

-$37.38 for the day, using $10 stakes. Now that's one heck of a losing strategy.

Yes it was in trial mode, but it felt just like I trading with real money - not so easy on the mind.

By the way I'm still a firm believer in Peter's suggestion that you can learn a lot by doing things randomly.

If anyone has any suggestion to tweak this strategy I'm happy to re-run the test and post the results.

I've uploaded an image of the stats for my '3-in-1' losing strategy

Regards,
Switesh

[Iron]

Hi Switesh

Next time, try offering to the queue. I bet you'll get closer to break even...

Jeff

[switesh]

Jeff, I did offer a price in my second and third attempt.

In the second attempt when I was backing top 4 runrs I asked for 1 tick above best market price - which is essentially offering a price. I did the same on the lay side as well (1 tick below best market price) in my third attempt when I backed top two rnrs and layed the next two rnrs.

When I offered a price some of the bets didn't get filled because the price moved rapidly into a 'potentially' profitable area. So overall offering a price din't make a big difference that I anticipated it would.
------------------------------------------------------------------------------------------------------------
So what am I going to do differently today ? Well I'm going to make a couple of changes:

1. Make the strategy more random, i.e. I'm going to use excel randnumber(0,2) to suggest 0 - "Back", 1 - "Lay", 2 - "No bet" for each of the top 4 runners in each race.
2. I'm going to use Implied Probability stake rather than use Fixed Stake.

The only thing I'm going to keep constant is that I will still play all of the top 4 runners in each race.

I'll be keen to see if I get a series of four 2's (1.186% probability) i.e. "No bet" for a race. That would be hilarious, cause excel's telling me "hey, there's Risk in the market - stay out".
The same probability would apply for all 4 "Back" or all 4 "Lay", or actually any of the other combinations as well.

Objective for today: To test if 'random selection' and 'probability staking' will deliver a significantly lesser loss compared to yesterday if it isn't closer to a breakeven result.

Regards,
Switesh

[herbie]

I have built a scalping bot which i can control how much it loses on a setting and wont win.... I havent got a scalping bot that can win yet... but hopefully very close

[switesh]

Here's the result for the Random strategy I ran on Monday.
Strategy: Randomly Back, Lay, or Do Nothing for top four runners in each race using excel randbetween(0,2) where 0 = "Back", 1 = "Lay, and 2 = "No Bet"

One race had all "No Bets" and 2 races had all "Lay".
I skipped the few evening races.

Result (attached):
Overall, an improved strike rate than my previous experiment and a lesser % loss on Market Turnover, but still a very drab strategy.

I will run another iteration of this strategy next week and the parameters I will be changing are: either cutting out losing positions manually (i.e don't allow losing positions to run for the full 6 minutes exposed to risk) or introduce trailing stop-loss.

Any suggestions welcome.

- Switesh

[Euler]

If you deploy strategies that take prices they will always lose in the long term because the market is pretty efficient and you will lose money on the spread.

[switesh]

If you deploy strategies that take prices they will always lose in the long term because the market is pretty efficient and you will lose money on the spread.

I implemented your suggestion for the third iteration of this random strategy. I did another randbetween(2,5) to offer price (ticks). So the bets (randomly generated) were placed anywhere between 2 to 5 ticks away from the center of the book.

Numerous bets didn't fill and as a result less money went through the market. But overall the result (attached) was much better than the second iteration - Very close to breakeven and close to 50% strike rate.

Oddly enough, Ffos Las (or 'Fast Loss' as you call it ) produced the best result.

In the next iteration I will try to cut out losing positions manually, but if I know myself well, I will only worsen the losing position.

Regards,
Switesh