Peter Nordsted commented on the previous post, by referring me to a comment on a similar observation made last October. Back then, Peter wrote:
To clear this up I do freelance writing for Matchbook where I am asked to concentrate on the Top televised matches for the Premier League and Champions League and I am asked to do a write up and finish with a recommended trade for that game.
The analysis I provide for Matchbook is based on a statistical view point but also contains my own subjective opinion.
The selections that are produced for Premier Betting are based on a situational model that has proved profitable over the past few seasons and contains no subjective opinion of my own whatsoever.
Indeed some of the selections go completely against my personal thoughts with the Everton v Newcastle game being a point in case.
For example 2 weeks ago the model as well as the Drawmaster (which is based on a similar model) suggested backing the draw in the Everton v West Brom game whereas my personal opinion would not have been to take this position on.
From an investment point of view I invest in the Premier Betting selections along with other selections in different leagues which the model suggests.
In the previous 2 seasons all of the Premier Betting picks were based on myself and Danny’s personal opinion but we decided that we wanted to share part of our profitable portfolio with members and this is how will go going forward.
I hope this clears matters up and just to reiterate although I co-head Premier Betting none of the selections given by Premier Betting contain any subjective opinion.
This approach has highs and lows and we will only be judged by results by the performance at the end of the season come what may.I suppose one benefit of advising Heads to one audience and Tails to another is that one group will be pleased, but the latter is not the paying Premier Betting Account subscribers this season.
Although Pete starts with "to clear this up", in fact, the confusion was compounded.
The Matchbook analyses is "based on a statistical view point but also contains my own subjective opinion."
The Premier Betting selections "are based on a situational model that has proved profitable over the past few seasons and contains no subjective opinion of my own whatsoever." What Peter means by "based on" is not clear. Either the model is objective or it is not.
So the question remains, even if you have a profitable (objective) model, (and the evidence suggests this is not the case), why would you diverge from this model and offer opposing (i.e. losing) advice to another audience?
One follow up comment from the time was this one:
Well first things first that clears up nothing. The explanation makes as much sense as having different selections for paying and nonpaying viewers. If Matchbook are happy to just have subjective views that's one thing, but why would you have them in the first place, if you have a successful situational model for paying subscribers (are you not convinced of its worth?)....ignoring your own businesses advice is pretty damning evidence of its quality.Analytical Spider offered this at the time too:
I like Pete and don't think it's an integrity issue as has been alleged but surely amongst the biggest gambling follies is going with subjective judgement bets. Unless you're betting on niche markets or lesser known markets I find it highly unlikely there are many if any serious bettors betting weekly on the big leagues, using their opinion and making long term profits. I've never met or heard of any and I've met a few having been a full-time trader/gambler for 10 years.
Given Pete's apparently got a successful data based strategy I find it a little strange he'd even want to offer subjective opinion (certainly on one like an over/under market in the prem). Personally I've found the longer you're successful at betting the more you realise how fallible opinion is (certain markets/situations excluded) and I'd tend to agree with the above comment that it doesn't reflect well on the quality of his model.I like Pete too. In all my dealings with him, he comes across as a very nice man, but most of his Tweets suggest selections are subjective:
I think the champions offer some value in the Man Utd v Liverpool game. See preview #MUFC #LFC"Think"? What price does the model have for Manchester United? "Some value"? How much edge does the market price offer?
We should be in for an entertaining affair when Spurs meet Arsenal tomorrow #THFC #AFCPeter uses 'entertaining' to mean 'lots of goals', but if the market expects lots of goals too, where is the value? What price does the model have for the Unders markets? How much edge does the market price offer?
I think there is value in the O2.5 goal line in the A Madrid v AC Milan #ECL"Think"? What price does the model have for the Unders markets? How much edge does the market price offer?
An 'objective situational model' will generate these numbers, and if it is profitable as Pete asserts, why not give those numbers and why contradict the profitable model with subjective opinions anyway?
Finally, there are contradictory Tweets of this nature:
Man Utd have a strong home record v Liverpool. Please see full betting preview #MUFC #LFC
Man Utd have only won 1 of last 5 matches hosting teams of Liverpool's stature #MUFC #LFCAs I have tried to explain before, the first error here is in confusing a club with a team.
Manchester United and Liverpool Football Clubs, like any football clubs, are not fixed entities. Their composition is constantly changing, often from game to game, certainly from month to month and season to season. Players come and go. Injuries, suspensions and tactics affect team line-ups. Managers come and go. Philosophies change. Players improve, players peak, and players decline.
Comparing a club against another club is almost always a waste of time, and the further back you are looking, the more futile the exercise is.
'Strong home record' is a very vague phrase anyway, and given Manchester United's recent history, it could probably be used for every home game this season, but any profitable model would recognise that Manchester United of 2014 are not the same Manchester United as 2013, 2012 or any recent seasons.
It may be a curiosity that Manchester United (the club) has a 'strong home record' (whatever that means) versus Liverpool (the club), but it's useless data (noise) to put into a model.
Who made a betting decision based on this information I saw yesterday?:
This is an extreme example admittedly, but if you understand why this information is, from a betting perspective, useless and dangerous, then it should be easy enough to extend that understanding to comprehending why comparing club versus club results is also useless from a betting perspective.
1 comment:
Considering you have about 4 separate entries in your own league table it's a bit rich to criticise Pete. I'm sure many people running systems will have 'overlaps' or multiple selections just because of the way the data is interpreted. Maybe Peter's brief from Matchbook sometimes means the result will be at odds with his own view and much like you cherry picking and highlighting whichever of your 'entries' is doing best Pete will do much the same.
Post a Comment