Monday 31 Dec

Happy New Year:


Here are the Year End figures and below the breakdown between Turf, A.W. & N.H. selected races.

Over the next few days I will post other breakdown figures - by Dutch Frame size, etc.
And try and outline a revamped blog for 2019.
If you look back through the blog you will see I set a benchmark for the returns of 35k, so another good result

So on the last day there was only 1 no: live selection. 

Columns AE - AQ
There were a couple of loses over the last few days and the final P/L figure of 41,939 (to the standard 486 stake) dropping slightly from a max. 42,326 on the 28th.
You can see from the screen shot below the Edge and strike rate figures there were 1711 live selections taken with114 painful loses.

Columns L - O
This is the inserted additional bets on Targeted favorites.
For inst taking only those returning > 95 to the 486 stake. The addition was 11,616 off 290 live bets placed.



Turf Races Only:
 

 A.W. Races Only:
 

N.H. Races Only:

The turf figures were down compared with most prior years and NH particularly good. There's always a sizeable gap, as would be expected if you go back to the race by race fav. percentage and Uplift percentage figures which have always been stronger for NH racing. (columns I&J int All races top screen shot)

This year the AW edge has outperformed the turf, I find this hard to explain when we had a wonderful summer of good stable going on the turf? 
The introduction of many more Nov. stakes and fewer maidens with revised entry requirements may have an explanation and I'm going to dig into the data on this.

Conversely the fav. Target bets on the AW actually showed a small loss on the >95 return!


So I have to get off and celebrate All the Best and more to come in the New Year.





I missed replying to quite a lot of emails during the summer. Firstly an apology and secondly don't let it put you off getting in contact in the future. I enjoy everyone's thoughts and experience of the racing game. I will try to respond in 2019. Perhaps I need to separate out an email for the blog.
My standard email is bombarded with absolute rubbish daily from the usual suspects, shysters, tipsters, trying to extract funds off punters, offering their opinions, cobbled up methods and bizarre claims to be mathematicians, laying on a hot beach, in the know, the Bookies enemy, etc, rather than being able to actually demonstrate how to beat the book long term.

Sunday 16 Sept

Top Class Curragh Card:

Four Group 1 races down the Curragh card today What a weekends racing its been.

Here is the first ~ 206k 5f sprint. Margin bet pays 65/486 or Target down if you've got your preferences. Yesterday my St Leger pick on the day Dee Ex Bee faded in the final half furlong to 4th place. But I still made money on the race with the Dutch bets ( My tipping is definitely not worth following ~ but then neither is anyone else's long term!)

On the RH Book value chart we see the need for the 5 runner Dutch Frame to clear the minimum 82.5% book. In the Shape of the Market graph LH side we have only 1 no: additional runner in the 'Field Envelope' (those I consider active in the market at the field size) 

The return has jumped a little showing 71 on the margin bet in this screen shot.

Toward the end of last week the YTD return tapped through 30k on the standard bets. Its consolidated over the weekend, hoping there's no draw down on the horizon back below the 30k.

You can see in the YTD figures at the top, the strike rate has clipped 93% with 1,248 bets placed live. In these were included 87 losing bets dumping back 42,282 of loses. If only we could avoid all or more pitfalls!

There were a lot of races eliminated yesterday leaving just seven live bets overall + 2no: additional Target bets. Not optimal live bet selection but did avoid what would have been the one losing bet in the 13;45 Listowel. In this race the Dutch frame was down to 2 no: runners in a strong market, but the 4th in the market just turned over the favorite, Joseph O'brien runner first time hurdling at 16/1.

Saturday 15 Sept

St Ledger Dutch Field

Here is the Six runner Dutch Frame for this years Ledger.
Most of the pundits are describing it as a very average year but personally I think the top six are very competitive this year. Gosden and Frankies late confirmation of a run with Lah Ti Dah has ripped the market.

The six runner dutch pays around 50 to the 485 stake. Screen shot shows my additional Target onto 3 runners against the Gosden favourite. I'm going to have a few quid ew on Dee Ex Bee on course. He was competing above 1 mile with good speed figures as a 2yo + second in the Derby, seems a good value ew punt against a fav. to prove the distance? Perhaps a bit more rain would have helped.   

The six runner Dutch Frame should have the depth in the field covered otherwise.

 

Sunday 2nd Sept

Third Quarter; Results Roundup

The consistent good racing conditions through the summer produced some great monthly returns.
August however reversed the trend with a large drawdown. This was coincidental with the first rainfall and change in going for weeks over the hot summer.
I'm always very loathe to ascribe any specific cause to a bunch of loses - the random variables are always a stronger effect than any other?
The change in going in Autumn would be the signal for many trainers to introduce horses they were saving for autumn campaigns. The York Ebor meeting is full of these every year.

 So the old chestnut Correlation does not imply Causation.

Fact is drawdowns eating profit are hard to bet through. If your not absolutely confident of your edge you will not see it through 99% of punters fail through a combination of lack of bankroll and loss of belief that their edge will return and overcome short term volatility.
Or of course they are simply kidding themselves that they have found a long term edge.


Here is the graph showing monthly returns 2015 to August 2018 and the monthly/cumulative actual figures.

Th end result was not a loss for august and at the beginning of September the losses were recovered.

Here are the results for the weekend 1st + 2nd September.
There were 24 no: potential selected races , but unfortunately they were filtered down to 9 no: live selections. As it turned out all 24No: races would have been wins! Far from always the case.

Five of the filtered races were Group races which are heavily filtered out  - strength in depth often justifies this Sometimes I take these races on placing manual bets after having done more detailed research on the field. Mainly because I like to be involved in the quality races.

P/L at the 2nd of Sept is 29,210 with a strike rate on the live bets of 92.89% So I'm pretty confident of getting through the 35k base target for another year.

 
I'll post some details of the August reverse in fortunes, dates etc. A disappointing April a further hard hit in August but a great summers return between.

Sat 14 April

A wide Open Grand National

I  can't tip the winner but can advise that you can Dutch every runner up to 50/1 and get a return. So a Dutch bet covering over 20 runners, Targeting profit on around 8 to 10 of them is the way I will watch the race!
This gives me a return of 100 to the 486 stake.
 
Still no racing ground for the flat season.........heavy going and no bets so far. 
A good flat card at Leopardstown today - but only 4 runners turn out for the 2000 Guineas Trial. So I'll continue to patiently wait and hope one of my Targeted runners in the National gets there.

Monday 4 April

First Quarter; Results Roundup

There was a lot of disruption in the miserable January + February weather conditions, not surprisingly there were not great monthly profits delivered. Always quiet months. And now back to abandoned meetings over the bank holiday.

Early March produced a tremendous run right through to a good Cheltenham, so put me back on Target for the quarter.

Here is the graph showing monthly returns 2015 to March 2018 and the monthly/cumulative actual figures.

There have been some good April runs in past years - here's hoping! 
So far there hasn't been a single flat race bet to place as the season fired up - All the meetings so far offering nothing better than Heavy going! 
There would have been 19 selected races so far, despite the heavy conditions we would have been only 1 no: losing race since the 24 March opening day. Would be good to get some decent racing ground and get the bets rolling on the turf?


Here are the stats YTD 2018:
Strike rate 91% off 451 Live race selections.
P/L 5,754 off the 486 unit stake.
Also the additional Target Bets have performed well with a P/L of 4,216. This of 212 additional Live Bets placed.

Going note:
The only races I eliminate due to going are turf races on Heavy.

Saturday 17 March

Cheltenham Roundup

There were seventeen selected races over the four days. Of the seventeen one would have been a losing bet - but its unlikely anyone would have been drawn into the Grd 1 Novice hurdle on Friday with 20 runners?

Shape of the Market filters left me with 11 no: live races and a return on the festival of @ 850 with the additional Target bets I took. 

I'm not a punter weighing in often these days. I had just two serious 'On the Nose' bets on favorites in a couple of the selected races. Footpad on Tuesday at evens and Altior on Wednesday at 11/10.

It must have been one of the easier years for punters to make money, I'm sure quite a number did and I must seem over cautious to those who did well! 

Hope you were one of them?

Friday 16 March

Cheltenham ~ Gold Cup day

Two out of the 4 no: potential Dutch races are live for today.

First up the Triumph Hurdle - this year unusually a tight market and only 9 no: runners lining up. Has the fav. frightened them off?

Here is the 5 runner Dutch Frame and also the additional fav. Target bet which I will be taking on Apples Shakira. I might extend the Target to include Alan Kings Redicean, which is attracting money, it doesn't reduce the return below 100 and two onside feels better in the Triumph!!

In the Gold Cup we have an extended Dutch Frame covering 9 no: runners.
I will probably Target four of them as per the second screen shot, rather than take the standard equal return Dutch. So Might Bite, Native River, Killultagh Vic & Road to Respect on side for me with cover off the other five.

 Good luck for Gold Cup day.

Thursday 15 March

Cheltenham ~ thursday

Good day for me yesterday. Sickening to see Ruby Walshaw down and a repeat leg injury.
Today there are 3 no: potential Dutch Bets.

• 15;30 Stayers Hurdle is a no bet.

 Leaving us the three.

• 6 runner Frame in the 13;30, 4 runner Frame in the 14;50 & 5 runner Frame in the 16;50.

The 14;50 Ryanair Gd1 chase is subject to a higher price Gap between the last in the 4 runner Dutch Frame and the Next in the market(currently 14~15 needs to increase before the off). Subject to a higher than 1 pt Price Gap currently showing this will be a live bet too.

Best bet today is the Grd 2 novice hurdle 16;50. The 5 runner frame + fav. Target Bet - best on the card.

Looks a tough punters day today? Good luck.

14;50 Update

Just about got the Price Gap near the off, so took the bet 
 

 Balko Des Flos late gamble right up to the off and turned over the fav. Un De Sceaux quite comfortably!

Wednesday 14 March

Cheltenham ~ wednesday

You'ver got to admire Ruby! Straight back from a bad leg break (4th time that leg!) and two winners on the first day. His hard drive finish on Bene Des Dieux in the OLBG mares hurdle to take the outsider with Apple's Jade well turned over kept me out of trouble on day 1.

Today 4 no: races qualify. The three runner 'Frame' in the Queen Mother chase 15;30 looks a particular strong bet - but confess I have taken Altior as a value priced fav. and one of only two bets I am chancing on the nose!


Here are the markets /Frame sizes from the 3 runner frame in the 15;30 up to a 7 runner frame (favorite hedged) in the 16;10.

 First day was very much a punters day - hope it holds up for Altior going in later. Both the second and third favs. are strong opposition leaving Altior at a relative value price. The sensible advice is probably stick with the Dutch Bet!!

Tuesday 13 March

Cheltenham ~ tuesday

First day of the festival has just the two live bet situations for Dutching today, from the five potential non-handicap races. (Good luck with the Cheltenham Handicaps!!)

Both the Champions Hurdle 15;30 and the OLBG Mares Hurdle 16;10 have strong favs. in Henderson's Buveur D'air and later Apple's Jade. Grade 1 racing to savour at last, neither fav. is a well clear proposition off most speed figures!

Plenty of factors in play with the first two days on the Old course and potential heavy going.

I will just be trying to get through day~1 without mishap and take the available Dutch Frames of 5 runners with Buveur D'Air and the tighter 3 runner Frame with Apple's Jade.

The Supreme Novice Hurdle doesn't qualify and Footpad's  market in the Arkle novice chase  14;10 is currently too tight although it may imrove towards the off - I would be taking the Target bet on Footpad if the market does ease?

Thursday 1st March

Wipe Out & Continuing

It's been terrible fare through Jan and Feb and v-hard to turn a buck!
To be honest I've done so much coughing and wheezing combined with the intermittent racing, my interest has been minimal.

First day of March and no racing at all across the UK, many meetings already abandoned into the weekend, with less than 2 weeks to the start of Cheltenham. I need the markets to settle down and deliver some consistent playable races.

Here's the state of play to the end of Feb with a couple of month end losses keeping the miserable trend going!

Monday 5 February

Neck and Neck

Watching the last of the evening card today ~ Here's the 7;15 Nov. Stakes at Wolv. tonight.

 The race offered up a three runner Dutch Frame for me on the margin bet profiting off the first two , break even on the third in the market.
I guess at least 75% of punters would have favoured one of the top three in the market? In fact why guess the Bk value of the Dutch bet was 90% - so 90% of the money was thereabouts if not punters.

Screen shot coming up to the line with Galactic Spirit getting half a length clear in the last few yards. But jock Atzeni persistently used his whip on the right side and barged the favorite while right alongside. Honestly a bit more of it and he would have lost the race to the fav. in second place.
So if you backed the winner was it luck or skill? It was 100% luck and if you don't think so you're kidding yourself.

 In 2014 a statistical sports betting site Pyckio was set up a huge hit monitoring sports tipsters and offering their records and selections. Joseph Buchdahl ~ an absolute must read sports stats. guy ~ offered a detailed analysis in 2015. His detailed conclusion over the 6,000+ tipsters registered covering over a million tips was that although 2138 had recorded some sort of profit only around 45 could be statistically proven to be outside the realm of luck. here is Josephs summary.

 Time to draw conclusions. If ever there was definitive evidence that almost all sports tipsters do almost nothing more than replicate chance, here it is. Of course, to re-iterate, none of this proves beyond doubt that no skill whatsoever is operating. This statistical analysis has nothing to say about skill, merely whether a performance looks something like that which could be predicted by chance. It is conceivable that amongst the 6,000+ tipsters there may be a small few who are genuinely doing what they do with more than just lady luck looking after them. Indeed the analysis of long-term records containing a 1,000 or more tips shows that, survivorship bias aside, some of them at least might possibly be doing something better than that predicted by chance alone. A bigger and better analysis of more tipsters that can properly take into account the described survivorship bias might reveal a more statistically significant profitable fat tail containing skilful tipsters. And Pyckio being the largest (and most transparent) tipster social network online at present is in the best position to find them. Nevertheless, we are clearly not talking about many as a proportion of the total population of people playing this game.


What have we got to deliver a long term edge Dutching our selected races?
a) Long Term stats. on the high performance of the first three in the market in each of these race type and the reliability of the market.
b) The Favorite-Longshot bias: operating in these betting markets offering a small percentage of value 'within' the top of the market group
c) The elimination of many random events, as per tonight's race stewards enquiry. Or falls, blocked in, unknown ground, distance etc, etc, the endless list.

 Then we accurately set the Dutch Frame to match the Shape of the Market in each race and filter out any races as best we can outside the envelope.

The favorite-longshot bias is an aspect of sports markets recognized since the 1950's Buchdahl and many other knowledgeable guys have covered it, both from the punters and the bookmakers influence.
One article that particularly drew me was a 2010 study, "Explaining the Favorite -Long shot Bias"' It's a freely downloadable paper registered on-line as IZA DP No: 4884 from two statisticians covering over 6.5 million runners in the USA.

I got home to an email box full of 'offers' last weekend, not just the usual tipsters with 50 or so results in their track record, but also bots and one that really caught me! An automated version of VDW ~ Van Der Wheil who mesmerized many readers of the Sporting Chronicle in the 1980's and the then editor Tony Peach. There was lots of theory, data which was v hard to collect in those days, mystique, but never quite proof of the end result? 

More quietly at the same time from a copy of  1987 'Braddocks' a racing bible of the period which was handed down to me. Here is page 18 drawing attention to 1985-86 previous season novice Hurdle races.

Just about 40% of the races taken by the fav. Currently this has actually risen to an average of 44% over the last 10yrs. Explained by the reduced field sizes we have now compared to 20-30 years ago.

Trust me these long term stats are a solid basis for success!

Sat 27 January

Race Selection; Gruss QPL

To match up on Race selection you need to set up the Quick Pick List in Gruss as per the screen shot.

Menu Bar > Options > Preferences > Quick Pick List /markets

Tick off the All UK win & All Ire Win in the LH auto load box.

Type in the following list in the exclusions box, highlighted RH side:- (the bottom 4 are redundant on latest version but can leave in ok)

PA
Hcap Hrd
Hcap Chs
NHF
Nursery
Hcap
Sell Stks
Without
Reverse
Forecast
 v

So simple List of 7 exclusions to generate correct Race Selection QPL.
Check the rest of the Preferences box for match. You need to close and re-open Gruss to retain this set up

This will produce (for Sat 27 Jan the 12 N.H. races + 3 A.W. races) the correct Race Selection List ready for export to excel.

The final screen shot is the 'Live Price' import from Gruss. Middle to RH side is the transfer to my own Library format. I leave the export step manual as I like to check the listing on a daily basis. From time to time Betfair insist on slightly changing , re-spacing a race type description, which then needs adding into the library.

Tuesday 23 January

Dutch Bet Configuration

Every time I see that monthly comparison graph in the previous post that losing 2016 September month irritates me to hell! Where did that come from??


Here are screen shots showing the Four bet configurations referred to around the blog.

 1. eq Liab
Equal return on all runners included in the Dutch Bet to a fixed Liability.
Stake Total 486 ~ Return 75.80

 
































 2. fav_Hedge
Return set to a 100 return on all runners except the favorite.
Balance of stake onto the fav. In this example 51.84 return off the favourite and 100 return off each of the other 3 runners included in the Dutch Bet.

 3. margin bet
In this bet  the return on the longest priced runner in the Dutch bet is set to zero with equal return off the remaining runners.
This bet configuration kicks in if the return is low (below 50) or to allow the Dutch Frame to stretch to an additional runner to achieve the minimum 82.5% Book Value. 
Three runners return 88.23. Stakes 250, 93, 84 & 59 

4. Target Bet
This bet configuration  (automated) sets the return off all runners in the Dutch bet to zero with all the profit targeted to the fav.

Of course placing the bet manually you can Target any runner or combination ~ if you are blessed with information straight from the horses mouth!
st runner as away over Xmas New Year watching the racing from the sunshine! 

Return on the fav 150.81. On the other 3 runners in the Dutch zero

Update: The above screen shots where not taken in a live bet situation and are overlayed on an early  market before it is fully formed.  They are only to outline the stakes and returns using the four different Dutch Bet configuarations. In the past some people have jumped to conclusion that on 4) Target Bets I was just moving to a single bet on the favourite. So the screen shot shows the stakes across all 4 no: runners included in the Dutch bet and the profit taken on the favourite only.


Adding the Live Screen shot for this race.
It ended up stretching to a 5 no: runner frame to cover the two equal priced  4th + 5th in the frame. I was surprised the bet didn't switch to a margin bet and push the reurn above 50 ~ I need to take a look at that.

fav' took the race, it qualified as a Target bet too but below the min. return of 95 I take them.