Losing strategies wanted

[Talkbet]

I reckon it lost about £10. Loads and loads of tiny wins but also a fair number of hefty losses.
Had to laugh at Le Tiss's comments about curry and roses though

[Shamrock]

I tried this about a year ago with excel for a month. I got in 60 seconds out with a lay on the horse with the most volumn, looked for two ticks greened and let the unmatched bets run. Never layed anything greather than 5.00.
Had some good days and had some bad days but the only thing I lost was a month of my life. I'll have a look for the spreadsheet but I think it was a profit of less than $10.
So my guess is a small profit of less than two quid.

Shamrock.

[Iron]

Hi RG

I disagree.

If my system were a case of 'enter at random with a stop of x ticks and an offset of y ticks', you'd get that result.

But with the approach I've outlined, if the market starts trending and you don't hit your 1 tick offset before it does, you could easily lose 20 ticks.

In my early days on Betfair, the market would go against me, and rather than take a small loss, I'd keep waiting and hoping for it to turn around. I'd think 'it can't carry on drifting forever', and inevitably my losses would get bigger and bigger! OK, there might eventually have been a reversal, but typically I ended worse off than I would have done had I just taken a small loss on the chin when the market started to move against me. So I know first hand that this is a losing strategy!

Jeff

Statistically, over the longer term , your loss should have been near enough to £11 - turnover less 5%.

rg

[rubysglory]

Hi Jeff. Looks like we agree to disagree. Such is the beauty of this forum.

rg

[Iron]

RG - Why agree to disagree? Why not try to get closer to the truth?

The horse racing markets, on the whole, have a propensity to trend. And if you enter a market at back at 3.0 and sit tight whilst the market drifts to 5.0 (something that happens all the time), you'll take a huge hit, if you're relying on 1 tick profits to pay for those huge losses. If you take 20 tick losses a few times in a day, you'll need quite a few 1 tick gains to compensate...

I think the mistake you're making as seeing the markets as purely rational - as constantly reflecting the true odds of a horse at any given time. If the market were purely rational, then extracting an edge would be a theoretical impossibility. And the irrationality of the market is confirmed by Betfair's own stats: http://promo.betfair.com/tactemails/camb1a.htm. When a market starts to trend, often everyone jumps onto the trend. Swing traders grab a piece of the action, and mug punters think 'the horse is steaming, so there must be a gamble on. Someone knows something!'. As a result, markets frequently trend for further than is rational.

Jeff

[rubysglory]

I agree to disagree Jeff because it much more polite than saying I think you are wrong ! Ooops I said it ! I think betfair's example data is flawed because of the timeframe ( 2 hours prior to the off ). I assume this is why Euler started his sample analysis four minutes out and exited at post time. The closer to the off the closer we come to achieving True Odds and market efficiency.

rg

[superfrank]

And the irrationality of the market is confirmed by Betfair's own stats: http://promo.betfair.com/tactemails/camb1a.htm.

I'd be very interested to see that analysis comparing the price at 4mins or 3mins to SP to determine a drifter or steamer.

[CaerMyrddin]

Running that strategy through many markets would make you lose between 7 to 10% of your stakes?

LeTiss's post was brilliant

[Euler]

And the irrationality of the market is confirmed by Betfair's own stats: http://promo.betfair.com/tactemails/camb1a.htm.

I'd be very interested to see that analysis comparing the price at 4mins or 3mins to SP to determine a drifter or steamer.

The problem with those stats is there is a bias in the market from two hours out to post time, so I'm not sure they are accurate? Also, do those figures include commission?

[Euler]

The numbers from yesterday.

-£6.48, it only had winners on 30.77% of races which is really, really low. Total turnover was £1936.83. So it was a loss of around 0.33% on turnover.

It wasn't a very efficient strategy from a trading perspective, but still didn't lose us that much.

[Iron]

I agree to disagree Jeff because it much more polite than saying I think you are wrong ! Ooops I said it !

There's nothing impolite about saying you disagree with someone.

I knew what you were saying, but this is an interesting issue, and worthy of exploration IMHO...

I think betfair's example data is flawed because of the timeframe ( 2 hours prior to the off ).

I don't see how that makes it flawed. It makes it limited in terms of what it tells you about steamers closer to the off, however.

The closer to the off the closer we come to achieving True Odds and market efficiency.

The market relating to that sample wasn't very efficient at the off!

What makes you think that a market created by a wide variety of people, with different skill levels and different reasons for being in the market, is likely to be efficient?

Jeff

[Iron]

Also, do those figures include commission?

I don't know.

BTW, I don't see why the same principle wouldn't apply to closer to the off. If anything, I suspect that laying steamers (and vice versa) offers more value if the move occurs in the 10 minutes prior to the off, as IMHO the market is likely to contain more traders, crowd-following mug punters and bookies laying off liabilities.

Have you done any research into steamers and drifters?

Jeff

[Iron]

The numbers from yesterday.

-£6.48, it only had winners on 30.77% of races which is really, really low.

I find that surprising. You'd have thought that, due to the market's latent volatility and the fact that 50% of the time you'll guess the direction of the trend correctly (if there is one), it would be at least 50%...

Total turnover was £1936.83. So it was a loss of around 0.33% on turnover.

If you had just entered the market at random, and had a preset offset of x ticks and a stop of y ticks, what would the ROI be?

Jeff

[PeterLe]

Jeff too many questions!!

Maybe, I'm interpreting wrongly....but I think what Peter has tried to show is a STARTING point for a strategy..its the first step in a 100 step process. However, even as a starting point, its not going to lose your bank. (There are a few posts on here that show betting at random with no stop loss will be about break even)

What you have to do now, is to build on it. Run the program and collect a decent sample rate and then look for patterns (and they will be there, if you look hard enough) of losing bets.

Then try to refine the strategy again and again.
It is rewarding both mentally and financially when you crack it, but you have to put the spade work in.
Just have a go! You can even do it in practice mode, so no downside all upside

regards
Peter

[Iron]

Peterle -

I think we're talking at cross purposes here!

I'm not devising a strategy. I'm just discussing how well my system performed at losing money, and how efficient the markets are.

But thanks for the tips.

Jeff

PS No such thing as asking too many questions!