harry_rag

Bet 12 [Rugby League] Australia,Saturday,10.35am
 
Parramatta to beat Rabbitohs @ 1.83 with Paddy Power 
Stake,6pts

I have a vague idea for an additional "value rating" to add to what we did for the Guineas races (though an individual could use it as well). At it's most basic, you could multiply your rating by the odds for that horse. The number might be a bit ungainly but you could do what I did and convert it into an index number where 100 is the average for that race so anything above 100 appears to offer the better value.
Say you have two 9/1 shots in the race, with speed ratings of 68 and 72. That gives you 680 and 720, with the potentially better value selection getting the higher number.
Or, say you have two horses rated at 70, one is 9/1 and one is 12/1. That gives you 700 and 910, again flagging the likely better bet.
I'm not suggesting this is a magic bullet but it could speed up the process we went through of mentally comparing the ratings to the prices on offer and picking out our best bet. A useful indicator to add into the mix perhaps.

Welcome Rob.
You branching out into stand up mate?  Only joking, they are indeed extremely knowledgeable and there's a good mix of tips, advice and theoretical discussions. They even let me take part.
Have a look at the footy threads as well, any input most welcome there, especially as focus turns towards the Euros.

Hi @Rob W. Welcome to the forum. The horse racing community are very modest here so they'll never blow their own trumpets but I can honestly say they're a cracking bunch with a superb knowledge base for the sport. You'll not only learn a lot from the staff and members here but you'll have a great time doing it too! 

Welcome @Rob W, you will find losts of info and tips on here, wishing you good luck.

I have a vague idea for an additional "value rating" to add to what we did for the Guineas races (though an individual could use it as well). At it's most basic, you could multiply your rating by the odds for that horse. The number might be a bit ungainly but you could do what I did and convert it into an index number where 100 is the average for that race so anything above 100 appears to offer the better value.
Say you have two 9/1 shots in the race, with speed ratings of 68 and 72. That gives you 680 and 720, with the potentially better value selection getting the higher number.
Or, say you have two horses rated at 70, one is 9/1 and one is 12/1. That gives you 700 and 910, again flagging the likely better bet.
I'm not suggesting this is a magic bullet but it could speed up the process we went through of mentally comparing the ratings to the prices on offer and picking out our best bet. A useful indicator to add into the mix perhaps.

To be fair I am looking at several angles here and this particular analysis was in response to the question posed by @Trotter.
For this analysis the selections were the top rated overall. This would be a starting point for further analysis. Hopefully as time progresses I will discover analysis based on the going that are more profitable.
All my work is with the intention of ending up with a pure system, my interpretive skills are not very good !
Speed figures are not a particularly precise measure, I suspect that they probably produce the wrong answer in most cases but they just need to be right a certain amount of the time for them to be valuable.
My main area of focus at the moment is to look at horses that have produced a speed figure in the top 50% of performances and then further analyse them based on a number of factors such as course, going, class, position,weight etc.
So as an example
Following horses in the top 50% on their next 5 runs gives the following results
Original run on good going, 6,063 wins from 46,673 runs, profit 1,262, AE 1.04
Official rating > 95, 2,049 from 17,925, profit 1,078, AE 1.06
Original run in class 1 to 3 races, 7,693 from 66,489, profit 613, AE 1.03
Fillies / Mares, 6,818 from 52,515, profit 422, AE 1.03
2yo, 1,466 from 8,718, profit 190, AE 1.04
I now need to look at these in more detail to arrive at a logical system that will hopefully prove profitable in my test data and then in real life !
 
 

For clarity, what are the criteria here? Backing the horse that was top rated on its last run subject to the minimum odds criteria? If so, is that top rated on the going or overall? Just after a reminder of where you’re up to.
Am I right in thinking you’re trying to arrive at a pure system approach where the selections pick themselves? I do think that speed ratings lend themselves more to an interpretive approach than an automated one (unless you can find a way to use them to price up every runner).
I can see a scenario where you’re backing a top rated 5/1 shot when there’s a 25/1 shot that’s rated only slightly lower which would be the obvious bet for anyone using their judgement. Or are you taking the odds into account in some way?

I've had a look at the data for the last 2 years, it proved a bit trickier than I inittially envisaged.
As you suggested, ignoring those at the top of the market will improve returns.
Ignoring those horses with forecast odds of 7/2 and below gives an AE of 1.05 with 1,117 from 9,558 runs with a profit of 296 points. Not a massive return but a good starting point.
 

Well done guys, it’s actually quicker to commiserate with Tony for not backing the winner rather than congratulate those of you who did!  It would have been a good week for me to have just surfed the wisdom of the crowd.
I’m curious re this one, it wouldn’t make much difference to the total return but wouldn’t there have been a dead heat settlement as there was a 5 way tie for 18th? I’d have expected the return to be based on 3 5ths of the stake. Be good to know if there are books out there that settle the markets “including ties” rather than “dead heat rules apply”.

I backed Boniface at 19/10 yesterday (void if not starting) and whilst that's shortened slightly he's bigger on the exchange and may not start. Wirtz should start and looks good at 3/1 with 365. Their 25% BB offer applies to the game so stick him in with Leverkusen win and whatever else you fancy for a decent tilt at that.
Leverkusen deservedly favoured but Scammaca is decidedly hot for Atalanta. He's scored 10 of their last 31 goals scoring in 9 out of the 17 games in which they scored them. Overall he's scored in 15/25 starts (including 3 braces) and scored 32.7% of Atalanta's goals in those games. I've taken 4.6 on the exchange.
Boniface has scored in 14/23 starts (also 3 braces) which represents 25.37% of Leverkusen's goals. I was happy with 19/10 yesterday but I'm unsure if he starts now with 3.4 available on the exchange.
WIrtz has scored in 12/36 starts and has 13.8% of their goals. I've taken 4.3 for him.
Might leave it there until the line ups are known.

So far I've had 5 points e/w on Italy outright at 18/1 and Lukaku top goalscorer at 28/1.
Italy are still 18/1 in one place and on the exchange (or if you can boost the 16/1 with Lads) Lukaku seems to be shortening; still 25/1 in places but a bit shorter on the exchange.
I read a preview on the main Golden Boot contenders and thought Lukaku wasn't bad at the price. Another that looks worth a second look is Dovbyk (Ukraine) at 3 figure odds but, so far, I haven't been able to get on at quite the best price or place terms.
I backed Italy when they won the last one based on Kevin Pullein tipping them in the RP. They might merit being bigger odds this time based on form but I still think the argument of backing one of the "usual suspects" at a bigger price holds water. There's so much luck in tournament football that qualifying and friendly form is of little relevance. There's a few who feel too short for me above Italy in the betting and plenty at bigger odds I can't see going all the way. At the price I'm happy to go with the Italians getting their act together when it matters (and, hopefully, enjoying a bit of the rub of the green).
Anyone else got any thoughts or had any bets yet?

That's because you should be doing 8x8x8x8 for 4 7/1 shots!
One most definitely would not! That was the point of the simple 6x6 etc. comparison. Adding numbers together has very different outcomes to multiplying them! Let's say you had the 8 selections in your 2 L15s one day, if you put all the 7/1 shots together you'd get a better return from that one if all 4 came in but it would be less likely to happen. If you want to have 2 L15s with similar prospects of a full house then you need to mix the prices up accordingly (which it sound like you generally do).
I'd say carry on as you are in terms of mixing up the odds but try and get your head round the concept of why two sets of numbers aren't the same when multiplied out just because they add up to the same. 
 

7x7x7x7 = 2,401. 7/1x7/1x7/1x7/1 =4,096 all 4 7/1 winners on a lucky 15 is 9x9x9x9 =6,561 ((knock the 1 off if it is a £1.00 for accuracy))
I have gone to great pains in the past to produce many explanations and a how to calculate a L15/31/63 post and posted them on this site it is a great pity that some people who are involved in the Lucky 15 thread didn't bother to read them properly because i specifically did them for the beginner. I tried searching for the table i posted but its like looking for a virgin girl in a brothel. Forget the odds, pick 4 horses that you think might win and as a rough guide if the odds add up to about 50 (in a discussion with the equaliser i dropped that to 30 to suit his fear of losing)  so at around 40-50 think about ew, any less down to 30 consider just adding an ew acca. below 30 forget the ew altogether Dead f'in simple.

That's because you should be doing 8x8x8x8 for 4 7/1 shots!
One most definitely would not! That was the point of the simple 6x6 etc. comparison. Adding numbers together has very different outcomes to multiplying them! Let's say you had the 8 selections in your 2 L15s one day, if you put all the 7/1 shots together you'd get a better return from that one if all 4 came in but it would be less likely to happen. If you want to have 2 L15s with similar prospects of a full house then you need to mix the prices up accordingly (which it sound like you generally do).
I'd say carry on as you are in terms of mixing up the odds but try and get your head round the concept of why two sets of numbers aren't the same when multiplied out just because they add up to the same. 
 

That's because you should be doing 8x8x8x8 for 4 7/1 shots!
One most definitely would not! That was the point of the simple 6x6 etc. comparison. Adding numbers together has very different outcomes to multiplying them! Let's say you had the 8 selections in your 2 L15s one day, if you put all the 7/1 shots together you'd get a better return from that one if all 4 came in but it would be less likely to happen. If you want to have 2 L15s with similar prospects of a full house then you need to mix the prices up accordingly (which it sound like you generally do).
I'd say carry on as you are in terms of mixing up the odds but try and get your head round the concept of why two sets of numbers aren't the same when multiplied out just because they add up to the same. 
 

That's because you should be doing 8x8x8x8 for 4 7/1 shots!
One most definitely would not! That was the point of the simple 6x6 etc. comparison. Adding numbers together has very different outcomes to multiplying them! Let's say you had the 8 selections in your 2 L15s one day, if you put all the 7/1 shots together you'd get a better return from that one if all 4 came in but it would be less likely to happen. If you want to have 2 L15s with similar prospects of a full house then you need to mix the prices up accordingly (which it sound like you generally do).
I'd say carry on as you are in terms of mixing up the odds but try and get your head round the concept of why two sets of numbers aren't the same when multiplied out just because they add up to the same. 
 

Haha many thanks for your detailed analysis.  My calculator is a Casio FC -100v Financial Consultant  so I don't know why it should give out wrong results.  BTW it shows 7x7x7x7 = 2401 not 4096 as your one does.
The point I was trying to make was whether mixing up the odds from a total number of odds of 28 in different combinations altered the chances of success.  One would think that they would be the same.
I didn't mean to confuse the issue over when it is a good time to only do a win L15 or an ew L15.  I had just lost @Zilzalian's notes that's all.  Sorry about that.
I don't have much time this week but I am all about fun with the L15 bets these days.  I like to do four of them when there are a number of meetings at 1.50 a pop is only risking 6 points for 16 selections.  I am trying to avoid favourite selections that are less than 3/1 with a view to getting on at 7/2 or above if the prices change.  I generally work through the shortest price selections upwards from A- D so I have a mixed bag of prices for each L15.  When looking at a race my idea is to select a horse at the biggest price that has the best chance of winning
Many thanks for your thoughts
 
 
 
 
 

You've gone an extraordinarily long way around the houses to arrive at a fairly basic mathematical conclusion here! 
6 x 6 = 36, 8 x 4 = 32, 2 x 10 = 20
3 pairs of numbers that all add up to 12, the ones that are closest together have the greatest product. That's how multiplication works.
I think you could do with getting a better calculator. Nothing too fancy, a £7 Casio job from Sainsburys would be fine. As it is, both your answers are wrong as you've got 1 too many zeroes after the decimal point. That said, why on earth did you do that calculation rather than just work out the accumulative odds? Four 7/1 shots gives you 4096 while your "mixed bag" gives you 2145. Surely comparing those numbers is easier and more relevant than the figures you've arrived at to multiple decimal places?
 
You may be right, it rings a bell. Similar to saying you'd want average odds of around 7/1. Perfectly sound as a rule of thumb. Personally I'd probably still be leaning towards the win only approach for 4 7/1 shots but each way would be more viable than for the mixed odds. I wouldn't be going e/w where the selections included a 2/1 shot but then I probably wouldn't bother with anything that short in a L15 full stop.
I genuinely can't see any logical connection between the conclusion you've drawn here and some guidance on when to favour win only over each way! If you compare two L15s the one with the lowest acca odds is the one that's most likely to see all 4 come in; the one with the lowest priced selection is the one that's most likely to give you at least a single. If you're going to do a couple of L15s there may be some logic to having a mixture of odds in both rather than all the shorties in one and all the longshots in the other. If you just fancy 4 7/1 shots then stick them in a L15. Don't go off looking for other selections for the sake of varying the odds. That would be letting the tail wag the dog.
I think you've gone down a bit of a rabbit hole here (something I'm prone to do myself at times). If you're going to do a couple of L15s then by all means mix the odds up a bit but worry about the selections first and don't worry if you're fancied selections all come in at similar odds sometimes.
And get a calculator that didn't come out of a Christmas cracker! 

Final Result:
Sad to say we managed to produce the worst overall result of all the divisions, we were obviously overcome by the occasion of being in the top tier!  We wiped out half our overall bank and only 3 managed to stay in individual profit for the season although most of us did find at least one winner. But it looks like only our 1st place winner will be returning to Division 1 for the next season.
3 small wins this week plus an eliminating no show gives us a change to the top 4.  @Sterphyle is our well deserved winner by some margin with 7 wins over the season, a mix of double and treble picks. @tonythepaint climbs up to 2nd place with a 5th win, and @Cauncie maintains his 3rd place.  @Jediknight was our other weekly winner and takes the 4th place prize although not in overall profit.   Well done to everyone for trying so hard all season, see you down in a lower Division next year.

On Sunday @black rabbit picked ZIGGYS PHOENIX at 66/ 1winner
On Monday @luckypants picked EVELYN'S PHOENIX at 33/1 winner
 
Perhaps we are over complicating matters, just pick a horse with PHOENIX in it's name !

YTD -£19.95
Doing 2 this week. Both £1ew
C. Kirk 60/1 B365, 5 places 1/4 odds.
D. McCarthy 50/1 WH 8 places 1/5 odds 
Good luck.
 

My picks this week

By my numbers Scheffler is well clear of the pack, he's ranked #1 on 6 of the 16 key metrics that I've focused on for this event.
...but I can't justify backing him at 3/1.
Morikawa has very strong numbers in all those key areas (bar putting: where he's 121st across the last 2 seasons in strokes gained putting) and he's as well placed as anyone to challenge him, especially coming off a strong showing (tied for 4th) at the PGA Championship.
But my biggest stake bet this week will be on Max Homa - who is comparable to Morikawa with his irons and Stokes Gained in general...but both putts & scrambles better - at a bigger price. It's a gamble, as his accuracy off the tee would need some improving this week and that's a tricky part of this course.
 

 

Best of luck Tony and Phil.
I like the look of two players this week, both with Bet365
Daniel Berger 100/1
Parker Coody 275/1

It can get really complicated when place terms come into play as well. Many years ago I did each way perms on the top nationalities markets in a Major and the shop hadn’t got a clue how to settle it. I didn’t know either but I knew they weren’t offering me enough. I ended up speaking to a guy in their back office who explained it really well so I’ve had a decent understanding of it since then.