Wednesday 11 September 2013

Improbable Is Not Impossible

Paul stopped by on the last post to say this:

Sorry for making such an uninspiring first post, but I am more a consumer than contributor being very new to the trading side of things (and damn do I have a long way to go!).
But I was hoping you might be able to indicate where John Walsh's NFL selections are to be found? The NFL is one sport where I feel a bit more confident, but (possibly due to not seeing what is right in front of me) I can't locate where his predictions are published - nor whether they are 'free' or a subs service?
Thanks for any help, and for the most entertaining blog which is a great help to a 'newcomer' such as me.
John Walsh's selections are available via Peter Nordsted's Trade on Sports newsletter:
If you would like to receive these picks on a weekly basis throughout the NFL season then you can do by subscribing to the Trade On sports Newsletter here
I believe they are free, although Peter's use of the word 'subscribe' implies otherwise, but Peter is a frequent visitor who will doubtless clarify the situation.

John finished off the NFL weekend with a nice winner at 2.65 as the Philadelphia Eagles comfortably beat NFC East rivals Washington Redskins and also had the Over 50.5 come in for a total profit of 2.15 points.
On the subject of traders being either unwilling or unable to share their loss size : win size ratio, Anonymous suggested another possibility:
Or maybe you're just overestimating your own importance to readers of the blog?
That category would be covered by the "Can't be bothered, or aren't interested" category, and that's all fine and dandy. I happen to like these kinds of things, but I understand not everyone loves their numbers as much as me. It remains an interesting topic though, and I really would like to see more data to confirm or refute my theory that the ratio for a perfect trader will in the long run be close to 1:1, with, of course, more winning markets than losing ones.

Mark Iverson stopped by after I had penned the above to add some data:
Only just had a chance to take a look at my numbers (since 2008)....
NFL -
Markets: 250
Ave Win: £86.15
Ave Loss: £83.10
S/R: 79.6%
In comparison....
Cricket (until end of Aug) -
Markets: 1199
Ave Win: £230.85
Ave Loss: £231.59
S/R: 84.24%
Interesting to see that your NBA results are as important to you as cricket is to mine.
Given Mark's experience with trading, his average win size slightly exceeding his average loss size in the NFL, and very close in cricket, is in my opinion, testament to his prowess as a trader, especially when you consider Mark's strike rate which is actually quite remarkable.

As for the comment that my NBA results are as important to me as cricket is to Mark, it doesn't take a rocket surgeon to calculate Mark's winnings from cricket, and they are some way ahead of my NBA winnings. I do have him beaten on the NFL though, and a few of those games were ones where we co-hosted a trading room and I remember Mark's approach being a little different to mine. Whereas Mark tapered to the point of little trading late in the game, where volatility becomes more extreme, my approach was, and still is, to be more involved later in a game.

A classic thread from 2007 has been resurrected on the Betfair Forum. The opening post by a 'kenniky' proposed a modified Fibonacci based progressive staking system on under / over 2.5 bets and added that after eight losing bets ...
"Naturally you have no profit...but it's impossible to lose 8 bets in a row"
As is always going to happen with these silly ideas, less than a month later (certainly ahead of schedule) the eight losers in a row came in and kenniky was never heard from again.

I saw a quote the other day:
When you are dead, you don't know you are dead. It is difficult only for the others. It is the same when you are stupid.   
The Conference prices have finally been updated at football-data.co.uk and the top (in the money) of the revised FTL table now looks like this:
Gundulf had a post on his blog about the FTL table, noting the similar strike rates for many entrants, and noting the differences in returns. It's too early in the season as he says, but it will be interesting to see how these numbers mature.

4 comments:

Anonymous said...

Is that the "Professional Sports Analyst" Mark Iverson? So over 7 years he's made £200K, just have to hope his football app is bringing in the big readies cos his trading isn't.

Bayes said...

I've posted on the Betfair forum, but I'll post here as your blog deserves some recognition. I visit all the time by the way, it's just I don't contribute very much to anything!

I'm waiting for my manual trading historical data from Betfair so I'll paste that later, but I have a couple of bots that trade.

Their figures over 3 months are:

Markets 1513, Win rate 50.9%, Win/Loss Ratio 0.815

Markets 2724, Win rate 57.8%, Win/Loss Ratio 1.115

The first bot operates in a fairly volatile two runner market, so I'm not surprised that it has a lower ratio.

I will be interested to see how my manual trading numbers compare.

Bayes said...

Ok, I've done my numbers for tennis. They date from Sept 2005, although I've been on Betfair since 2001. I went full time June 2007 so most of my tennis trading before 2005 was fairly insignificant.


Markets: 5374 Win rate 62.4% Win/Loss ratio 0.998

I thought if I looked at purely pre 2010 my figures would've been a bit more like your own but they are not that different:

Markets 2570 Win rate 62.6% Win/Loss ratio 0.961

It does at least confirm what I thought about my manual trading that I have become better at staying at market. The other consequence of this is that I spend less time cowering behind the sofa.

Pete Nordsted said...

Cassini,

John Walsh charges me a fee for his NFL picks and I then provide these picks to readers of my newsletter which is a £15 per month subscription.

He also provides picks during the NHL season

Pete Nordsted