Thursday, August 28, 2014

When to believe your eyes (or at least the data table)

When trying to obtain data for use in any kind of forecast you are faced with any number of questions with varying degrees of complication. Is the data from a reliable source? Do we have enough of it? How should it be interpreted? When is it stable enough to be relied upon?

That latter question is where this post will focus. Generally, more data is better than less data. Nate Silver fans’ ears will prick up at that simplistic statement as his excellent book The Signal and the Noise is full of instances where too much data can cloud our judgement, but for our purposes let’s say that when trying to judge the quality of a team you’d prefer to have data for 10 games rather than 5 (data from every game in Liverpool history would start to be too ‘noisy’ as Bill Shankly or Ian Rush have little bearing on the current crop of players).

After two games last season, Everton had amassed 42 shots (26 SiB) giving them a crazy 21(13) average. While the team played well the rest of the way, their totals of 14 shots and 8 SiB were considerably below that initial surge, which could have led to a couple of panic buys as managers sought to get 'coverage' of 'must own' teams. Looking at this season to date, what should we make of West Ham's 35 shots (22 SiB) or Swansea's 15(7) efforts?

More learned statisticians will likely be able to analyse this question with more certainty and skill, but for our purposes, we are just looking for a quick guideline as to when we can believe what the data is showing us.

For simplicity, I have simply plotted each teams’ average shot totals (both in total and those only in the box) for the season against the rolling average on a gameweek-by-gameweek basis. These lines will obviously converge as the season progresses but the speed at which this happens is less obvious. The data is plotted below with some quick analysis below the chart:

By GW6, of the forty team/location pairs (20 teams each at home/away), 34 see their rolling average within just two shots of their final season total. Thus if at that the point in the season a team had an average shot total of 10, we’d expect with some certainty that they would finish the season with between 8-12. The only notable departures were Sunderland at home, who fell from three strong performances (strangely including Arsenal and Liverpool) and a 21 shot average to just 14 on the season and then Liverpool at home, who improved throughout the year, taking their 15 shot average through GW6 to 21 by the time they fell just short of a title bid.

It's dangerous to draw too many conclusions through six weeks, especially when further splitting the data in home/away games but ultimately we can't wait until we're absolutely sure (if that day ever even arrives) as decisions on transfers need to be made sooner rather than later. Still, six weeks feels like a good benchmark to start taking things a bit more seriously and putting some weight behind any big revisions to impressions you had coming into the year. From memory of Silver's aforementioned book, I think this is something akin to Bayesian inference, where our initial hypothesis should be impacted by new data but to varying degrees based on how strong our initial opinion was. Thus, if you loved Alexis Sanchez coming into the season, his somewhat disappointing three shots in two games should move the needle less than David Nugent's zero SiB, as you were probably less sure on the Leicester man's prospects initially (though even there, two weeks is probably too early to panic unless you've seen any real issues with his or Leicester's gameplan).

After being away for a few months I'm sure everyone is thrilled to read a piece which basically tells you what you already knew, but hey, I had limited data to play with while on a recent flight and this is what I managed to cobble together. This also ties in well with my plan to launch the new graphics and forecast tables right around the GW6 mark. Next up is some actual analysis of the new season. 

Tuesday, August 19, 2014

The more things change . . .

Personally, I've just wrapped up one of the best and almost certainly most important years of my life, having got married, travelled to four continents, finally got a new job and bought a house. For this blog though, the results have been less promising. I considered charting the quality of content here with my life developments or perhaps dousing the fire of my own work but given that I'm still travelling I'll stick to simple words for now.

Long story short, I've had priorities which have trumped this blog which meant that (a) the weekly content has suffered (and stopped at the end of the last season) and (b) I totally ignored this year's pre season activities. My policy has always been to only post things which are worth reading so I didn't ever want to mail anything in with out of date data or banal narratives. I wasn't sure I could make a quality preseason guide so I didn't and I wasn't sure if I'd even be back for this year. Now it's started though, I got that familiar buzz on opening weekend - even if my team was assembled the night before - and so I've come to the conclusion I'm not yet ready to walk away.

There will be a couple of changes though. First, I'm hoping to move to a more 'graphic' based site which I'll hopefully host on a new site that allows for a bit more flexibility. Second, I won't aim to put out weekly lineup lessons or 'fanning the flames' pieces which take up masses of time and in all honestly become repetitive for you to read and me to write (no, you shouldn't buy the 19 year old right back who played once but scored with his only shot of his life). I will however continue to post written pieces where a particular player needs attention or where a new concept/trend arises.

There are a lot of good sites around which cover player fitness, team news and what I'll call 'standard' reporting and while I've never tried to offer great depth in those areas, I'm abandoning that area entirely now. I know less about the weekly ups and downs of football than most of you probably do as I'm simply not plugged into it 24/7 thanks to living in Canada. I no longer default to Sky Sports News as my background noise and I don't discuss Rooney's hamstring in the elevator at work anymore. The problem with this approach is that when data tables show Stevan Jovetic as the best forecasted player for a given week despite knowing that there's a 99% chance he won't play, many people get confused/annoyed and complain. You can't please everyone though and there likely won't be comments on the new site anyway (I'm always on Twitter though for any fairer comments or queries).

So the plan for the next couple of weeks is to get the new graphics completed and launch the new site. That should nicely coincide with the time when we have some somewhat useful data (~GW5). In the mean time, I'll start getting back into the swing of things by highlighting some promising new players and offering caution to those whose early success looks unsustainable (basically a prolonged fanning the flames piece).

I've just realised that given my absence there could be no one reading this but if you are, thanks for sticking with me and I hope and I can reward that loyalty with a couple of useful tips in the coming season.

Friday, February 14, 2014

The cost of caution: delaying transfers

Striking the balance between risk and reward is a difficult task in fantasy sports and getting the wrong mix can result in wasted points and eventual failure. Perhaps this is me projecting my own views onto the analysis, but I feel that sometimes those of us who are more inclined to make decisions based on data are generally more concerned with the risk of too much action: such as paying ill-advised four-point transfers on a weekly basis or captaining players based on nothing more than a hunch. I often speak - as do those more learned than I - about the risk of relying on small samples and the need to regress anomalous results back to some form of mean to avoid overreacting to chance events, and these are indeed good virtues which will generally lead to fantasy success.

What I want to talk about here though is another type of risk: the risk of inaction, specifically as it pertains to making transfers. I noted last week that my team is valued at a measly 103m while the league's top teams are pushing into the 110m, or even 115m range. Given my moderate success in the past and the constant turnover of players, I've argued that chasing cash is not necessarily a worthy strategy in of itself and a quick glance at the game's "most valuable" teams will highlight that team value does not always correlate to fantasy points (with a major factor being the transfer prices paid by those teams, of course). I am not fully backing up on that approach as "chasing money" still seems like a strategy that wouldn't pay dividends by itself, as by definition, to get the most gains you need to be selecting players that everyone else is also picking, and those people are generally not the ideal role models for rationale behaviour. What does need to be looked at though, is the idea of making the transfers you want to make, but doing so as early as possible in the gameweek (h/t to reader CDI for raising this question).

At this point, I'm sure many of you are astonished that I haven't been through this before, but alas, I am indeed that foolish. My position has always been that I'd prefer to lose out on the odd 0.1m rise by making late transfers if that means avoiding injuries to players I've just signed. What I've failed to properly appreciate though is the surprisingly high cost of delaying your transfer by just a week. I believe this is something akin to what is often defined as loss aversion, particularly in the sense that signing a player on a Sunday night who then gets injured in training that week stings so badly, that it feels much worse than it is (the cost being, at worst, a four point to get rid of him again). If I delay action, I only lose something I never had anyway - that potential rise in price - but by delaying I ensure that I avoid wasting a transfer: something I tangibly 'own'.

I know reading about other people's fantasy teams is the very definition of mundane, but without copying the data for every player in the league, I've had to draw the line for the below analysis somewhere, so I've decided to focus on the transfers made by my own team this year. The methodology is to compare the lowest price I could have paid for a player with the highest for a given week, which will generally be the price paid if you wait until the last minute to make your move. For example, following his fourth goal in three weeks in GW6, Aaron Ramsey was obviously due for another price rise. If you made the move to grab him as soon as the GW6 fixtures finished, you'd have paid the closing price from GW6 of 6.2m. If however, you waited until right before the GW7 deadline, you'd now find that the same player cost 6.5m. I know this is obvious and self evident, but what is surprising to me is that this issue doesn't just apply to 'obvious' examples like Ramsey. Below are all the purchases made for my team this season, excluding those during the wildcard week (even I'm smart enough to make wildcard transfers on day one of the new gameweek):


Though this analysis will slightly overestimate my idiocy as on occasion I did make transfers earlier in the week, it still shows a striking impact in that delaying making these moves cost 2.3m on the purchase side and 1.5m on the sale side (this impact is lowered by the way sales price is calculated, but still, it's a material difference). Remember that this is purely a product of timing and I haven't changed my policy as to who to target at all, yet I've left something like 3.0m on the table purely to guard against the risk of injury. What's really surprising to me is that over half of my transfers were impacted by this effect, not just those following a week where a player produces a huge points haul thus triggering a stampede of transfers.

Assessing the real risk of injury is harder, as potentially I changed my transfer target based on injuries which happened during the week. Though I can recall a few anecdotal instances of avoiding injuries due to delaying these moves, I again can't help but wonder if I'm putting too much emphasis on those occasions as they stood out to me as "justifying" my position. I guess if there's mid-week European action, it might still pay to wait on transfers but otherwise it seems that being cautious can lead to a serious loss of funds in the latter stages of the season and right now I would happily give up 8-12 points to correct injured transferees earlier in the year if that meant I could have another three or four million pounds to spend.

To some of you this has probably been as enlightening as a long prose explaining how the earth actually isn't flat, but hopefully my late arrival to the common sense party will help illuminate reality for a few others.

Saturday, February 8, 2014

Can Sturridge replace Suarez?

I've noted a couple of times this week that I'm personally looking to make changes to my side given budgetary constraints and I assume at least some of you are in the same (or similar) position. I'd suggest the most common solution is to go with a "stars and scrubs" strategy, holding the star players like Suarez and Aguero (when healthy) while trying to fill your squad with a smattering of low priced picks who you believe you can squeeze enough value from to give you the edge. Generally, this strategy can work quite well as (a) you're better off with two players who score 8 points and 4 points than two who score 6 due to the ability to captain the star player, and (b) the game tends to leave a number of players undervalued due to a sudden increase in playing time or player improvement.

However, at this stage in the season, I am concerned about the lack of budget options. The best budget options have been bid up to mid level prices and thus if you want to hold onto the likes of Suarez you need to take a serious risk somewhere else, either backing sporadic starters like Januzaj or Gnabry, or pinning your hopes on one of the new arrivals to the league. This piece is not arguing against that strategy - as I'd suggest there remains enough promise at the budget end to make it work - rather it will try and establish how much you loose by making a downgrade at the top end of your team and what that can buy you in return.

I'm going to focus on the comparison of Luis Suarez and Daniel Sturridge, but you can make a similar comparison to Negredo/Aguero, van Persie/Rooney or Hazard/Oscar. Indeed, I will try and visit some of those pairings in future posts, but for now, let's look at the Liverpool duo.

Below is a snapshot of the two players' performances to date. I have only included games where they both played as (a) this is the reality of the current situation and I don't want to distort data with shots earned in a different position to the one currently occupied, and (b) if either player were to get injured, the argument becomes moot:

Note that the "expected" part of the above is based on the expectation of goals and assists based on their actual shot totals (rather than what the model suggests as a forecast).

So in all we have Suarez scoring two more goals and three more assists which equates to around 20 extra points. That's a pretty significant number, especially if you assume you'll be captaining the player every other week or so. The current iteration of the model is less bullish on this gap and projects just a 10 point gap for the remainder of the year, the difference largely being due to the fact that a projection model is never going to forecast Suarez to have a 10 shot game, as was the case in GW11. The degree to which you feel this kind of model underestimates Suarez will likely be instructive on how you interpret the rest of this post.

The current price difference is 3.6m which is also significant, although that number is greatly reduced for anyone who's held Suarez for a long time (I hold him at 12.1m for example, which gives a premium of just 2.3m). It's impossible to give a complete list of options for what that money can buy you, but here's a few ideas to put the numbers into context:

Lambert > Negredo +2.8m : Forecast gain of 25 points for the season (16 points using regressed rates)
Wilshere > Ozil +3.7m: Forecast gain of 9 points (16 points regressed)
Mirallas > Silva +1.9: Forecast gain of 21 points (16 points regressed)

It's coincidence that all the above options came out with a regressed 16 point gain but I guess it's a decent benchmark to sum up the kind of haul we can expect. Depending on how you feel about the model's view of Suarez, there seems to be an argument here to making the move to Sturridge and using that money to upgrade elsewhere. The captaincy issue is worth considering, though if you'd upgrading from a mid range option to an elite player, you will likely consider captaining them once or twice too, which would reduce that impact.

If your team is swimming in cash, I'm not sure this move makes sense. Suarez is essentially owned by everyone who's paying attention so you're opening yourself up to a huge risk by not owning him and I wouldn't advise making that play just to upgrade your first bench slot or to grab a slightly better 'keeper pair. But, if like me, you are struggling to even to put together a strong eleven, it looks like you might be able to squeeze more value out of Sturridge and another top flight player than Suarez and a mid level option. A final complication for those who've held Suarez for a while is that once you sell, you might not be able to afford to bring him back so this would be one mistake that literally cannot be rectified. This is about as hard of a decision as it gets, but I can't help but feel that continued inaction for those who have a low value team is going to lead to slow reduction in mini-league leads.