Laying horses in play 35% lower odds than starting price

"3 min before start I lay every horse roughly 35% lower odds than sp"

How do you know SP 3 mins before the off?  Or do you just guess it'll be about the same as the prevailing price at that time?

I think you've almost certainly just struck lucky.

Gruss give the forcast SP

re automation, you could do most on gruss/excel, but you will never know the exact off time for the 3 mins so you would just have to settle for 3 mins b4 scheduled off. 

I think you can do something re approximate race times and in play time for your cancellation half way through the race, but not 100% sure.  Don't think it can be done with straight formulas tho

Beat The OverRound  .......had a thread on laying IR.....Might be useful to you if you can check this out.

If it is working so well........why broadcast it on here?

System hari kari on the forum. Why do people do this?

Have to echo above two comments, although at times you feel like you want an opinion as to whether a system is likely a longer term bet, or just hit lucky...

No one reads this forum anymore anyway do they?

Perhaps you could post a 'golden goose' on here without scratching the actual market!

trouble with bets like that is you get caught laying all the bigger price horses that get an easy lead, it will only take a few meetings that

favour front runners and you will be in trouble because you will only be laying the few horses that are up with the pace,at least you have

won a bit now so you can give it a longer test  and even if you lose it back(which i think you will in the end)it wont be costing you nothing

to find out

Most of the matched bets are presumably in-running? (Can't be too many horses whose odds plunge 35% in the last three minutes before the start). So you've no need to guess the SPs, you could wait until the race actually starts before placing the bets...

The IR bets will lose in the long run I've tried something similar.

I did a thread about it. mostly ignored. yes it works.

You're not going to make money like this over the long term. However - I'd still suggest that you learn to automate it and let it run because that will get you going in robot world. The first Bots I ever made did something along these lines, they were done very simply with Gruss and Excel as someone suggested above. That's how I started to learn programming and to understand how the markets work. I have a much more sophisticated operation now but I still make a living laying horses in the run so I haven't moved so far.

Gruss with Excel is an ideal way to start because there's no programming to do - you can just put formulae into spreadsheet cells and off you go, Gruss does all the background stuff like logging in,selecting markets and so on that would be tricky to do without a little programming knowledge.
From there there's an easy progression into writing VBA instead of formulae, then you ditch the spreadsheet and use the COM and eventually you move on to writing free-standing applications from scratch. It's all easy steps and you can make money at every stage.

Ultimately you're going to need a better strategy, (what you've got now is a guess and some hope) but for the time being you can work with that because as long as you keep your stakes small you're not going to lose much and you can start to collect data and develop new ideas. My tips for any beginner botters are as follows:

1: You need to place a lot of bets, and I mean a LOT. Your returns will be measured in pence per transaction, not pounds.

2: You need to collect a lot of data - really a LOT. Variance is huge in betting so it's barely worth analysing a set with fewer than 10,000 lines and you're going to want a couple of hundred thousand before you can start to feel any confidence about your conclusions. Obviously at the beginning you don't have a big data set so you just have to make do with what you've got as it grows, but the important thing is not to draw false conclusions from limited data - don't project your winnings forwards (or your losses) and DON'T SCALE based on limited data.

3: Analyse - but DON'T FILTER. You can extract a winning group of bets from any random set of zero value bets by putting in criteria to filter out the losers. That's not how you do it. Look for reasons not trends.

4: Never leave bets sitting. You bet on the basis of information, if your bets are left sitting then your counter-party is betting with better (newer) information than you. You've put yourself at an inherent disadvantage.

5: Things will go wrong, so limit the amount it's technically possible for you to loose. Weird things happen from time to time that can trip up the best planned programs (even spread sheets) so try to program in absolute limits to what a Bot can stake and keep your account balance low so that even in the worst case you can't loose your shirt.

6: Don't stress, and don't hurry. If you keep your bets small and make as many bets as you can then you won't do worse than losing at your commission rate. You've got 4k to start with so you can do that for quite a while.

7: The only stake you need for now and a long time into the future is £2. Maximum data, minimum loss.

8: The above are good tips, but they're far from being the best ones I could give, no-one is EVER going to tell you the real secrets. They are never mentioned in forums, never sold in e-books and never discussed between anyone, even in return for money. The people who really know how to do it aren't going to tell you. I get messages from people who want to "discuss strategy" all the time - what do you think I tell them? Nothing. You have to work out the good stuff for yourself, and if you ever manage it you won't tell anyone either.

Follow all this advice and you still probably won't win - have a go though. It's fun if you like that sort of thing and you never know.

Every system needs some logic to beat the edge so
where is it?
Will lose eventually or before.

Surely the logic is that they are shortening too far in these cases, though I agree you need to ask why it's too far.

It assumes the SP is good overall estimate and the IR price is an overreaction.

use betangel to use automatic rather than the manual rubbish and keep stakes low and monitor over a long period,
keep to the same stratagy throughout and then evaluate over a long period
finnally only lay in  2 mile plus races preferably 3 mile plus.
never leave bets till end.
you will more often lay the winner you want a few horses matched preferably.

From memory, BTO's strategy was to lay selected horses, not the field, for 10% less than his projected sp
and cancel unmatched bets somewhere around 10 seconds after the off. Acorns etc..
He claimed a high success rate with this method.
He seems to have disappeared off the radar for some time.
DFC might have a better memory than mine!

LAY IN INCREMENTS OF 15 TICKS BELOW BOOKIES SP IN RUNNING BUT CANCEL 10-30 SECONDS AFTER THE OFF DEPENDING ON LENGTH OF RACE.

Hmmm, explains the traffic.  Will set mine to 34%......

I'm going 33%.

Now you're just spoiling it for everyone....

I tried this sort of thing once and then watched the races where I was matched.  It was pretty clear I was matched most often when the horse had a red hot start by someone who knew what they were doing. Where my lay was successful often it was pretty sobering to see how close they came to winning.  I was not getting value.  Occasionally I was matched for no good reason - a mistake or someone trading out but mostly the backers took my bet for very good reasons.  If you combined some form reading with this and laid selectively  it might work.  Watch a few replays of races you were matched in
- before you call it a winning system !

I posted this elsewhere but it seems relevant to this thread so posting it here:

I used to use an active spreadsheet that took readings 12 minutes before the off.

Then, if anything moved in 25% (i.e. 4.00 to 3.20), I laid it.

If it moved out and another hit 25%, it then backed the first one and laid the second.

etc.

Depending on field size (don't remember exactly), it was the 'third' one that hit 25% that usually won and only if it went further down than 30%.

Theory was:

a) if a fave, if the fave lost in the previous race, usually fave backers chasing.

b) first two were usually headlines on the day (jockey, trainer winning)

c) heavy lays on another horse initially shifted market

d) third was truer money, usually last minute 'get on' rather than steady movement in.

Parameters changed.  This was five, possibly six years ago, I believe. 

Not sure it would work now since several horses moving 25% these days.

------

This was years ago so not sure how it would work against the markets now.

It as also 10/12 minutes before with the attitude to trade out before the off.

Someone I worked for also gave me a piece of knowledge for all my betting which makes a difference and that is to classify a horse on its style of running and then couple this with the jockey's (i.e. Jamie "I love to come from behind like Fallon used to" Spencer).

Simply knowing which horses are Front Runners can be an edge on in-running.

Aug 16, 2014 -- 9:25AM, zooot wrote:

I tried this sort of thing once and then watched the races where I was matched.  It was pretty clear I was matched most often when the horse had a red hot start by someone who knew what they were doing. Where my lay was successful often it was pretty sobering to see how close they came to winning.  I was not getting value.  Occasionally I was matched for no good reason - a mistake or someone trading out but mostly the backers took my bet for very good reasons.  If you combined some form reading with this and laid selectively  it might work.  Watch a few replays of races you were matched in - before you call it a winning system !

And this is very true as well and is a general rule in betting.

If you think linear and just enter the same amount without assessing the race closely, you won't win.  A matter of tinkering the amounts down and sometimes up depending on confidence, depth of field, all the conditions. 

It is time consuming but better for the blood pressure and hair colour.

Wow I completely forgot about this post. Thanks for all the replies. Some good comments, some ridiculous ones.
I have continued doing this manual about 3 nights a week but i am canceling out of the bet about 1/3rd way through the race. I have had ups and downs clearly but am continuing to show a steady profit. I did drop my stakes a bit and am going with a system of $3000 kitty=$33 bets, $4000 kitty=$44 bets ect. I am upto $8765 right now but I have also been doing some small normal betting and up about $500 in that.

I cant respond to everyone but kenilworth, I would imagine the logic here is that 1/3rd or earlier through a race I am happy to take on horses 35% lower that sp. There seems to be value there. I have found the shorter races are easier to profit. I tend to get matches on 4 horses in 10 seconds or less on 1100m sprints and then close out. The longer races are hard to get more than one horse matched.

Anyway I will keep you all posted.

PS I have noticed there is so often 50-1 - 200-1 horses that get out to massive leads in the long 4000m plus races that get backed way way in to 4-1. Of course they never win but I really dont understand it. I have been so close to laying them for huge in-play but stopped myself. Does anyone know why this is? There was a 55-1 shot which got a big lead in england last night, backed into 4-1 and lost by 200m? Do these ever get up?

You can always check up IN RUNNING LOW ODDS and HIGH ODDS of wwinners and losers usind TIMEFORM on BF ....link below...

http://form.timeform.betfair.com/

how is this progressing now

Hi guys another way is to keep the fav, and Set Lay in running all the  Others @ $3.00 to $2.50 Inplay .
You might match 3 Others ..Just a Idea

I have tried laying the field at anything from 1.31 to 4.2 depending on the size of the field. Seems to work better at tracks that are not covered by ATR. I wouldn't think there is much turnover on Australian/American and South African racing, I never know which one's will go IR

@big aitch

"I never know which one's will go IR"

How come?  It's clearly shown both on the website & the API?