Yes, the main Arsenal one I read is arseblog. Other than that, its general info blogs like soccerbyives or theoffsiderules.

I'd be interested in reading a blog you created though.

I was wondering the same thing actually. Most footy I've stumbled across are either teams ones or blog networks. I haven't come across something like ball don't lie for any football league.

Your blog definitely sounds like an interesting idea. What kind of stats are you going to be using and where are you gonna get them from? There's definitely a dearth of football related statistics easily available on the internet.

I was thinking about setting up some kind of website/blog with stories about some 'unkown' - as in unknown worldwide/continental - talent. There would be several writers/bloggers all from diferent countries and writing about the talent in their country. All in English. Didn't think further then that really but thought it would be fun to set up something like that.

confirm. the only thing it had going for it was capacity, and that's been surpassed by so many other clubs now.

st andrews was a dump when i went there, but they've developed the rest of the ground now, rather than just half of it so idk

turd moor is obviously awful

would definatly read. would probably be quite popular if you have enough time to invest in it. not sure where you going to get all your stats for analysis as i cant find many. admitadly iv probably spent less than 5 mins looking.

replicating stuff from that hockey site would be nice

Right now I am using what's called the poisson model to predict outcomes. It takes goals for, goals against, and team schedules and converts them to an attacking and defensive factor for each team as well as (in the version I use) a leaguewide home-field-advantage factor (plus one for neutral sites if I'm dealing with competitions where there are such games). I then plug these into the poisson formula for any pair of teams I want and it gives the probability for each scoreline. These probabilities can be used to simply make predictions for a given week, or I can run simulations to look at the chance of finishing in a certain place in the league or winning a cup competition.

There are some problems with the Poisson model but it's been used before in soccer (in academic research papers I've come across) and hockey and does a pretty good job. It's a lot simpler than other models that I've come across and AFAIK there isn't a model out there that doesn't suffer from the same problems. Using it would give me the ability to be quick enough making predictions that I could pretty quickly analyze any league I can find a list of results for. For a while, the bulk of my articles will be based on this.

I've written two actual articles so far. Both discuss World Cup qualifying from CONCACAF. I give estimates for the percentage chance that each team makes the finals in South Africa, look at how important the next match is for each team and talk about how much the 2-1 win for Mexico over the US mattered for each of those teams. My intention is to carry out similar analysis for the other confederations.

I'm late to the party, but I had planned to write a preview for the Primera where I use last year's results and my subjective analysis of summer transfers to look at how I think each team will do. Everything takes longer than you think so unless I work on nothing but that I don't think I'll have it done before the season starts this weekend. I'll probably post it anyway sometime next week. Depending on how much time I have and what sources are available for the other leagues, I may do something similar for the EPL, Serie A, Budesliga etc. but probably not since it would be weird to do that 10 matches into the season. As I said, once things get going I'll have a weekly preview for the top leagues (and I suppose others that readers request). I don't want to use last year's results so that'll have to wait until enough matches have been played for the predictions to be reasonably accurate.

Mixed in with all that I'd like to discuss stats themselves. Sometime fairly soon I'll write an article on the model I'm using and discuss issues it has. As I said, it does a pretty good job but it's not perfect. Because of assumptions that aren't valid, it misestimates the likelihood of certain scorelines happening. I'm working on an improved, which I'll definitely discuss if I can sort through the fairly complicated equations well enough to make it useful. I also want to look at how much team stats like corners, fouls, bookings, shots on goal and time of possession matter. I tend to be a traditionalist so my gut reaction is that they don't matter at all, but that's definitely something that can be looked into.

I'll also do more random stuff like discuss whether the EPL or La Liga is better, whether offense or defense is more important (the answer may be different for different competitions, I'm not sure), whether some teams are built for league or cup competitions, and cliches like a goal just before half time being extra important and of course whether a 2-0 lead is actually dangerous. I have ADHD so random thoughts pop into my head all the time. If this thing takes off and I get enough readers to make it worth doing for a while then I don't think I'll run out of these ideas, particularly if people send me ideas or questions they've thought of.

For individual stats, it's tough. The obvious place to start is +/- (roughly speaking team goals for minus team goals against when you are on the ice/pitch). Unfortunately, it's flawed even in hockey and significantly worse in soccer because the top players play so much. The stat, at least in the superior adjusted versions, are relative stats that essentially boil down to how your team does when you are on the ice versus on the bench. Due to the subbing system in hockey, even the best players play less than half a game. In soccer the top players play 35+ matches per league campaign and tend to play the full 90 minutes. Despite the obvious issues, it still is probably worthwhile at least as a starting point. I still need to learn database programming, but once I do it should be easy to generate so I might as well throw it in.

I'm also working on an individual stat that is a serious improvement over +/- and I think would work really well for both soccer and hockey, but the math is even more complicated than the team stuff I'm working on and I can't do anything with it until I have improved at programming so I don't expect to have it ready any time soon.

WOW that was long. I have been thinking about doing this for a while and now seems like a good time to start it.

Sound good jared. You want to do this in on the 'traditional' blog way or a website?

I'm going to start it out as a blog. If it goes well and I get good enough at database programming to where I will have some stats and whatnot that people would want to sort through themselves (like you see at behindthenet for example) then I'll go with a website.

Starting a blog being free and apparently easy to set up is a big benefit as well since it's far from certain that it turns into anything with enough interest that I'll even want to keep working on it for a while. This allows me to test the waters without spending money or a huge amount of time. It seems like the kind of thing where anything could happen from it being huge and effectively a full-time job (extremely unlikely but within the realm of possibility, hopefully I somehow improve in poker if that happens since these things apparently don't make money) to it completely fizzling to the point where I might as well just post my stuff here since 2+2ers are the only people reading it anyway. I'd hate to commit a lot of resources and have the lower end of that happen, which is more likely than not probably.

I'll definitely read everything you do, for what that's worth. Sounds fascinating and like something I would love to do if I had the time.

Blog is up: http://analyticalfootball.blogspot.com/

I'd appreciate any comments or suggestions on anything from format to the articles themselves.

wow, pretty sick. will have a proper look at it in a while

edit: I'm a tard.

Very nice. I've just started to go through some of the posts. Formatting looks fine, but I'm reading the posts through an rss reader so it my perspective's skewed. I'm looking forward to reading what you're gonna be putting out.

Looks goods. Will read and comment in a few hours.

I am making a database at the moment where I can fill in some numbers from the games and all kind of different stats roll out. Not sure where I will use it for. A blog, a website, just because it is fun. Will def use for sports betting tho if I can create some usefull numbers/stats.

Jared the analysis is awesome, however I didn't read everything, sort of skimmed. Kind of wish this was about something other than CONCACAF, but thats whatever. I love the odds you crunched showing the teams odds of advancing.

Let me know if you need any help gathering data, I don't know enough about poisson models to help in that regard, but I can gather stuff for ya.

are you models for gambling? I take it the poisson model isn't used for betting over/unders? (empty netters, overtime goals not being the "same" as a regulation goal) ... but I guess you could just bet regulation lines.

I fooled around with a Poisson model for predicting 1st period goals but I've never really built it to a point where I'd want to back up the money truck on it.

what is your hockey blog url?

Certainly could be. The academic papers I've read took the Poisson model and used it to bet on games with positive returns. It gives a probability of each possible scoreline so one could use that for over/under. It does over or underestimate probabilities for certain scores, so that should be taken into account I guess.

I personally have never bet on sports other than the odd bet with a friend for fun, but that's because I've never had a system and could see myself having problems betting too much without one. I may use what I've done for that, I'm not sure.

I don't want the blog to be explicitly about gambling, but I could definitely see sports bettors being interested since a lot of it will be giving odds for stuff.

Don't have one. I've thought about starting one with very similar stuff. It's a lot tougher there though because so much more has been done. If I'm able to work out some more stats on my own it would be pretty interesting for the hockey nerds I think so I'd start a website or blog with that as well.

Rangers supporters begin to count the cost of Murray's legacy of debt

finally a Scottish journo tells it like it is. times are going to be tough in the near future for the Rainjurs :S

Jared,
a question that could be a potential subject for your blog.
Why on earth does the MLS have a salary cap?

We could really put the hurt on them with our 4th title in 5 years this season.

Ah, a good question that boils down to an overreaction after the USL folded. Good topic for discussion none the less

I've been saying it for a few years now, that Rangers (and Celtic too) are going to come back to the rest due toi their lack of funds/over-spending, still be better teams of course, but the gap will be a lot closer, and a real challenge could develop from outside the top 2.

Seems that day is now growing very near.

USL folded? Do you mean NASL?

Horizon, the primary reason is that MLS doesn't want teams to fail because they are spending more than they can afford.

In reality, MLS should be allowing teams that can afford to spend, to ****ing spend.

****ing England running so well...Modric ruled out of the rest of Croatia's qualifying and depending on his recovery, perhaps even the playoff if Croatia finishes 2nd. ;ladfskjlawef right before Wembley too...

Alright atmosphere but not a great ground.

Fratton park biggest dump of the premiership. I found the walkers stadium 10x better.

Worst atmosphere that I've been to is at craven cottage.