(BLOG) Laying the Lame: Betting against the odds favorites

[greenmark]

Fugazi At last, my efforts have been recognized, which means it wasn't all for nothing Stay tuned for updates, tomorrow I'm preparing a big post to answer everyone's questions.

P.S.: I hope I haven't offended anyone. If I have, I apologize.

I can't wait for your update tomorrow. You haven't offended anyone BTW. But it is apologise, not apologize.

[Dogfather]

As per tradition, yesterday's results:

4 wins
1 loss

Final betting report. I did not include my very first prediction, as it involved handicap race bets and a combination of tracks and grades that I did not use further on. I start from the second prediction (September 4th) where my strategy remained unchanged, except for the very last ones where I paid more attention to comments from previous races (this did not significantly affect the win rate):

Total races: 39
Wins: 31
losses: 8
Win rate: 79.49%

Starting from the beginning of this topic, I'm getting a lot of messages like "show the odds" "BSP?" "No one will accept a bet much lower than BSP." My strategy is based on market inefficiency before the final stage (your beloved BSP), but you keep writing this! It's one thing to use market inefficiency at a certain point, and another to "beat" BSP, which essentially means being the world's best analyst for greyhound races. But you know what? I accept this challenge. I will change my strategy and try to beat BSP. Most likely there will be fewer predictions (I expect 1-3 per day) and more work for me. Just so you understand, to provide the standard 5 predictions, I analyzed around 100 races (including the previous runs of the dogs).

Unfortunately, I won't be able to provide both my standard forecasts and "BSP forecasts" - it's too much work (and I'm also trying to get into horse racing). Therefore, for the foreseeable future, there will only be "BSP forecasts". I hope everyone will be happy with this now.

A small clarification: since I don't know in advance what the BSP will be, let's agree that I will bet up to odds of 3.1 (but I will consider it as a bet at 3, since the difference is small, and I'm confident that BSP 3.1 can be taken a bit earlier at 3). Any races above that, we'll consider that I'm skipping. I will take the BSP information from here: https://www.timeform.com/greyhound-racing

[Anbell]

You dont need to beat BSP, but you do need to consider price. (BSP is a reasonable proxy for price for most purposes.)

[Dogfather]

Over the weekend, I thought about how to change my strategy to beat BSP. I did a lot of backtesting; the results weren't exactly thrilling, but I think I'm starting to understand something. My goal is a 70% win rate. I'll start with a bankroll of 100 units, with each bet being 1 unit. Let's get started!

Romford

11:52 (R4) Grade: (A2) Distance: 400m Duchess Harbour

12:42 (R7) Grade: (A3) Distance: 400m Tea Boy Tony

Monmore

15:19 (R4) Grade: (A3) Distance: 480m Swift Super

Sunderland

Grade: (A3) Distance: 450m Diegos Penny

Yarmouth

20:47 (R10) Grade: (A3) Distance: 462m Grapevine

P.S.: If there are too many dogs with a BSP over 3.1, I'll have to adjust how I release the forecasts, perhaps providing them about 15 minutes before the race starts.

[Anbell]

Over the weekend, I thought about how to change my strategy to beat BSP. I did a lot of backtesting; the results weren't exactly thrilling, but I think I'm starting to understand something. My goal is a 70% win rate. I'll start with a bankroll of 100 units, with each bet being 1 unit. Let's get started!

Romford

11:52 (R4) Grade: (A2) Distance: 400m Duchess Harbour

12:42 (R7) Grade: (A3) Distance: 400m Tea Boy Tony

Monmore

15:19 (R4) Grade: (A3) Distance: 480m Swift Super

Sunderland

Grade: (A3) Distance: 450m Diegos Penny

Yarmouth

20:47 (R10) Grade: (A3) Distance: 462m Grapevine

P.S.: If there are too many dogs with a BSP over 3.1, I'll have to adjust how I release the forecasts, perhaps providing them about 15 minutes before the race starts.

I think we're getting somewhere, but having a goal of a 70% win rate is still only half of the equation. Everyone here could easily get a 70% win rate in any and every market on betfair. Win rate is only half of the equation. Price is the other half.

You dont need to provide 15 mins before the race. You could provide your proposed lays whenever you want, but the bets / pnl / backtests only trigger if the BSP is below your nominated threshold. And then you can review yesterday to determine profitability. And the next day you add those bets to your record, and so on. After a month you'll have 150 or so bets and a record to judge whether your approach is profitable.

[Euler]

You need to lay at an average price of less than 3.35 to have any chance of making a profit with a 70% strike rate. This is why price is so important.

As stated above you can pretty much create any win rate you like, it's the price that will make you money.

[Dogfather]

Over the weekend, I thought about how to change my strategy to beat BSP. I did a lot of backtesting; the results weren't exactly thrilling, but I think I'm starting to understand something. My goal is a 70% win rate. I'll start with a bankroll of 100 units, with each bet being 1 unit. Let's get started!

Romford

11:52 (R4) Grade: (A2) Distance: 400m Duchess Harbour

12:42 (R7) Grade: (A3) Distance: 400m Tea Boy Tony

Monmore

15:19 (R4) Grade: (A3) Distance: 480m Swift Super

Sunderland

Grade: (A3) Distance: 450m Diegos Penny

Yarmouth

20:47 (R10) Grade: (A3) Distance: 462m Grapevine

P.S.: If there are too many dogs with a BSP over 3.1, I'll have to adjust how I release the forecasts, perhaps providing them about 15 minutes before the race starts.

I think we're getting somewhere, but having a goal of a 70% win rate is still only half of the equation. Everyone here could easily get a 70% win rate in any and every market on betfair. Win rate is only half of the equation. Price is the other half.

You dont need to provide 15 mins before the race. You could provide your proposed lays whenever you want, but the bets / pnl / backtests only trigger if the BSP is below your nominated threshold. And then you can review yesterday to determine profitability. And the next day you add those bets to your record, and so on. After a month you'll have 150 or so bets and a record to judge whether your approach is profitable.

I want to achieve a 70% win rate by placing bets with BSP of less than 3.1, but this is a very challenging task... I thought about making predictions 15 minutes before if, say, 4 out of 5 predictions are above 3.1 (meaning those bets would be skipped). Of course, it would be more convenient for me to just keep track of my statistics and post the results on a forum every 10 bets or so, but in that case, no one would believe me

P.S.: It's possible that this whole idea won't work out at all, and I'll quickly realize that I can't beat the BSP.

[Anbell]

Over the weekend, I thought about how to change my strategy to beat BSP. I did a lot of backtesting; the results weren't exactly thrilling, but I think I'm starting to understand something. My goal is a 70% win rate. I'll start with a bankroll of 100 units, with each bet being 1 unit. Let's get started!

Romford

11:52 (R4) Grade: (A2) Distance: 400m Duchess Harbour

12:42 (R7) Grade: (A3) Distance: 400m Tea Boy Tony

Monmore

15:19 (R4) Grade: (A3) Distance: 480m Swift Super

Sunderland

Grade: (A3) Distance: 450m Diegos Penny

Yarmouth

20:47 (R10) Grade: (A3) Distance: 462m Grapevine

P.S.: If there are too many dogs with a BSP over 3.1, I'll have to adjust how I release the forecasts, perhaps providing them about 15 minutes before the race starts.

I think we're getting somewhere, but having a goal of a 70% win rate is still only half of the equation. Everyone here could easily get a 70% win rate in any and every market on betfair. Win rate is only half of the equation. Price is the other half.

You dont need to provide 15 mins before the race. You could provide your proposed lays whenever you want, but the bets / pnl / backtests only trigger if the BSP is below your nominated threshold. And then you can review yesterday to determine profitability. And the next day you add those bets to your record, and so on. After a month you'll have 150 or so bets and a record to judge whether your approach is profitable.

I want to achieve a 70% win rate by placing bets with BSP of less than 3.1, but this is a very challenging task... I thought about making predictions 15 minutes before if, say, 4 out of 5 predictions are above 3.1 (meaning those bets would be skipped). Of course, it would be more convenient for me to just keep track of my statistics and post the results on a forum every 10 bets or so, but in that case, no one would believe me

P.S.: It's possible that this whole idea won't work out at all, and I'll quickly realize that I can't beat the BSP.

It is very simple. You can do both. You can post the predictions in advance, and the results in retrospect. You're not reinventing the wheel here.

[Anbell]

You can even put all of your recommendations in a public spreadsheet if you want feedback

[Kai]

P.S.: It's possible that this whole idea won't work out at all, and I'll quickly realize that I can't beat the BSP.

Have to stay positive OP, much can be learned even from a failed attempt. If it doesn't work for you the process/methodology feedback from others (that already made it work) can still be useful to others.

That's what the forum is about, sharing methodologies that do work for us. The strategy behind the strategy if you will.

[Anbell]

P.S.: It's possible that this whole idea won't work out at all, and I'll quickly realize that I can't beat the BSP.

Have to stay positive OP, much can be learned even from a failed attempt. If it doesn't work for you the process/methodology feedback from others (that already made it work) can still be useful to others.

That's what the forum is about, sharing methodologies that do work for us.

True, but the attempts need to be tested in the real world. That's the only reason that any of us are here. We can learn a lot from the process, and we can test ideas vs the market, and we can test the logic here, but the ideas have to live or die by whether or not they are profitable in certain environments.

[Kai]

P.S.: It's possible that this whole idea won't work out at all, and I'll quickly realize that I can't beat the BSP.

Have to stay positive OP, much can be learned even from a failed attempt. If it doesn't work for you the process/methodology feedback from others (that already made it work) can still be useful to others.

That's what the forum is about, sharing methodologies that do work for us. The strategy behind the strategy if you will.

True, but the attempts need to be tested in the real world. That's the only reason that any of us are here. We can learn a lot from the process, and we can test ideas vs the market, and we can test the logic here, but the ideas have to live or die by whether or not they are profitable in certain environments.

Ye, goes without saying he needs an account of his own like I said on page one, not much use paper trading. I left some feedback in his inbox on day 1 on how pros deal with these situations, hopefully that helps.

[Dogfather]

Yesterday, all my bets won (except for Diegos Penny, who didn't compete yesterday - I must have included him by mistake). However, these wins don't count towards my record because the BSP was over 3.1. It seems that simply posting predictions for the entire day in the morning won't work. I'll need to analyze the races closer to their start times. Hopefully, this will help me avoid doing pointless work.

[jimibt]

Yesterday, all my bets won (except for Diegos Penny, who didn't compete yesterday - I must have included him by mistake). However, these wins don't count towards my record because the BSP was over 3.1. It seems that simply posting predictions for the entire day in the morning won't work. I'll need to analyze the races closer to their start times. Hopefully, this will help me avoid doing pointless work.

this is all good tbh.. by analysing the issues, you are drawing closer to having a testable strategy. as you say, it may fail or it may bump along on breakeven. however, it will mean that you've considered all the options and are fully prepared for live markets.

[sionascaig]

Identifying a possible false fav early still may still be worthwhile.. Either the market picks up on it & you don't get a match or it doesn't & game on...

Either way, BA can post your selections (at value price) and if you get matched all good..