Dies on its arse-- ...
 
Notifications
Clear all

Dies on its arse-- night before system  

  RSS

Avatar
(@satkinson_uk)
Trusted Member
Joined: 3 years ago
Posts: 53
15/10/2020 9:11 pm  

Ok- looking- the system test has tied on it's arse. I think a there are at least 3 reasons for this: 

 

@mickmcormack has gone quite= legend

people are very worried of prices going-understandable

people want to put up selection but not their methd- being made a monkey of isn't nice

3rd  and 1/2- hmmm, maybe the strategies available already! and their success

 

I am going to use a cobination of my own Time Decay basd card (this is my "cmon let me know" card) - Monte carlo of the card--- human analyis of the results.....Engage in a useful way and we can discuss-how we improve it

standards card analysis- monte carlo

Consider VW when I consider the % high enough (sorry, not witholding anything, just have not nailed a numbr to the mast- will start off with a "drop off")

Review PR raw score (market has een taken into account earlier)

Onl going to consider pace in flat races with TRUE- logic being that with the others, atings could become distorted

I may be, due to being busy- not post everyday- I will post my bets. I may not always post my rationale- if questions are asked I would like to try and answer-- but- single dad of 2- may not be very speedy coming back to you.

 

I ay also be a clever sod and post later- you can decide if I am a prick or if I am genuine. I am going to say I won't mislead--ultimately- other things in my  life take precedent !

 

PS- I'm not a grumpy person! 🙂 

 

 

OH-- you are all wondering- £££££?

 

From 1st September, until now, I made a £1000 profit in me betfair account-- and sorry again -Sometimes I have done 80/20; some odds attracted me more to 70/30----sraight win and  some e/w doubles...

So my strategy to followo is yeee hah- don't follow

 

 


Quote
Avatar
(@michael-clarke)
Noble Member Customer
Joined: 5 years ago
Posts: 1133
16/10/2020 12:31 am  

Yeah, not sure what happened to Mick, he was the only one on here who seemed to make the ratings work consistently.

I think Andrew struggled to find an edge using the ratings.

I did OK for a while but then there were various issues with the data that stopped me in my tracks.

The data is now better but I have struggled to find an edge with the ratings that perform better than my basic form analysis.

I will put up a NH system on here starting November including the system rules.

Hopefully we can reinvigorate the systems test thread, it was very helpful when it was running.


ReplyQuote
Avatar
(@andrewp)
Honorable Member
Joined: 2 years ago
Posts: 527
16/10/2020 7:36 am  

I think Andrew struggled to find an edge using the ratings.

I did OK for a while but then there were various issues with the data that stopped me in my tracks.

The quality of the historic data was an issue for me. I wasn't confident it was giving me a firm basis for building a model. Also, I already subscribe to a formbook which contains ratings I've been using for years and am comfortable with. It provides very good (but not perfect) data, is cheaper, offers discounts for winning competitions and doesn't charge for days when there is no racing (great during lockdown). I don't get one to one advice although there are some profitable free systems.

I still struggle to find an edge. 😊 

Mick had a rare ability to pick winners consistently at decent prices long before the race. Others have shown this too but over much shorter periods so more difficult to gauge. I'm hopeless at it. I prefer to operate at the 'off' where I have all the information and can adjust my earlier bets and proposed bets if necessary.

 


ReplyQuote
Avatar
(@mikeywilding)
Member Admin
Joined: 11 years ago
Posts: 2475
16/10/2020 9:33 am  

Looking forward to joining in @satkinson_uk

Follow the Race Advisor on social media on Facebook, YouTube, Twitter and Instagram. You can also find my personal accounts on Instagram and Facebook
.


ReplyQuote
Avatar
(@michael-clarke)
Noble Member Customer
Joined: 5 years ago
Posts: 1133
16/10/2020 11:45 am  
Posted by: @andrewp

I prefer to operate at the 'off' where I have all the information and can adjust my earlier bets and proposed bets if necessary.

My successful systems seem to be based on using only a small amount of information, I find that the more data you have the more difficult it is to arrive at a conclusion.

The RA ratings are an example, there are hundreds of them but I haven't been able to work out which are useful or not so I have gone back to just using my basic data.

A study at Princeton University called it an "illusion of knowledge", it gives people more confidence but not more predictive ability.

A lot of the data is "noise", not information.


ReplyQuote
Avatar
(@andrewp)
Honorable Member
Joined: 2 years ago
Posts: 527
16/10/2020 1:40 pm  

I really meant 'more market information' at the off Michael. This is when everything that can be known is known. For instance, at this point I know about non-runners, I know how the market has changed on every horse, i know the weather and going conditions. Market prices close to the off are critical to some models. I can also trade-out, lay-off the drifters or simply adjust my stakes. Very time consuming though unless you automate.

I quite agree about having too many factors in a rule-based approach. Half a dozen max should do. As for working out which ones to use, I've found statistical tests (correlation and regression) and A/E a good starting point. Then it's a case of pick the strongest, add another. If it improves things, keep it, if not ditch it. Add another and so on.

This gets you to a stable strike rate and sometimes a small edge at BSP, then it's all about price.


ReplyQuote
Avatar
(@satkinson_uk)
Trusted Member
Joined: 3 years ago
Posts: 53
17/10/2020 1:43 pm  

Ok- here goes

 

Catterick 3.10

Redrosezorro; Suzi's Connoisseur; Economix Crisis---- Dutch bet

 

3.45- Muscika 80/20 (4 places on betfair)

 

4.20- Splinter; Gleties; Bollin Joan- dutch

 

4.50- three saints bay; shawaamekh; erich bloch - dutch

 


ReplyQuote
Avatar
(@andrewp)
Honorable Member
Joined: 2 years ago
Posts: 527
18/10/2020 10:27 am  

Great start Stuart. To BSP

Actual 3

Expected 1.55

A/E 1.94


ReplyQuote
Share:
Back to top button
Close