Case Study in Goalkeeper Rotation: 2014/15 in Review

Struggling to decide on your fantasy football goalkeeper?? Have a read through this guest post from your friend and mine Mr. @FPLWalt. One of the most genuine folk I’ve ever had the pleasure of never meeting – Please drop Walt a follow on twitter – well recommended!

Roll on the new Season!

R @FantasyYIRMA

OH THE PAIN!

“NOT AGAIN!!” I scream at no one.  It’s May 2, 2015 and I have two starting goalkeepers. I play Kelvin Davis of Southampton away to Sunderland. Southampton lose 2-1 and Davis gives me 2 points. Later that day, my other keeper, Boaz Myhill, also plays away but in the dreaded and seemingly hazardous confines of Old Trafford, where Manchester United are fighting to secure their top 4 finish.  Myhill keeps a clean sheet, saves a penalty, and nabs 3 bonus points for a GW total of 17 points.  I watch another close H2H matchup and 15 HUGE potential FPL points slip through my Kelvin Davis sized fingers as I scream out and promise yet again that I will never ever ever rotate keepers for as long as I live; or at least until I’m able to calm down and have some time to look into the raw data that goes along with my emotional rollercoaster that is the FPL season. And that… is what the off season is for.

So, as the pain of the recently finished season dissipates and I gain my wits back, I put myself to this selfish task of trying to take the emotion out of things and seek data-backed logic and rationale for that dreaded question I ask myself every season and for which I still do not seem to have a firm answer. Do I want to rotate goalkeepers or not?  Is this perfect Home and Home rotation strategy that the Dr. Evils of the FA tempt us with really the way to go?

And so the research commences with a number of different thoughts pounding through my brain:

“Are goalkeeper points important enough to even dig into?”

“Do I really want to take all this time to do research, because you know once I start I will research it to death and try to look at every angle?”

And the dreaded….OH MY what if I do all of this research, spend multiple hours putting it together into the form of an article, and I end up with nothing meaningful to share with the FPL managers out there burdening themselves with the same question I am? Please make that not happen.

I take a few deep breaths and dive in, because at the end of the day, this is a question that will bother me in August as I stare at that screen, lost in my FPL mindwarp, knowing that the team I field for GW1 will set the tone for the coming season.  Goalkeepers may be an afterthought for most, but I want to make a decision that I feel good about when I hit that CONFIRM button, and even if my decision ends up being horribly wrong, I will know that it was the decision that made sense to me at the time based on hard work and the sweat and toil of feeling like I did what I could to back my decision up.

So Here’s What I Did

I went back and looked at the 2014/15 statistics trying to see if managers who used a Home and Home rotation strategy were rewarded by starting with and sticking to this strategy, or whether they felt the way I did on May 2nd; screaming at the thin air and losing a crapload of points along the way.

I made a number of assumptions:

1) FPL managers choosing to rotate GKs on a perfect Home/Away rotation basis do not want to spend more than 9.5m on their two keepers.

2) For the purposes of my research I will look strictly at H/H rotation and will consider the GK at home to be the keeper on the field for each week.

3) Nobody wants to waste a free transfer on a goalkeeper. Am I right or am I right?

2014/15 Home/Away Rotation Options Worth Considering

So when the 2014/15 Premier League fixture schedule was released, perfect home/away rotation pairings worth considering (total outlay of 9.5m max) were Aston Villa/West Brom (9.0m),Burnley/Hull (9.5m), Leicester/Southampton (9.5m) and Newcastle/Sunderland (9.0m).

Being that Newcastle/Sunderland are also one of this season’s perfect rotating pairs as highlighted by our friends at @AllThingsFPL and one at the forefront of many an FPL manager’s 2015/16 planning considerations, I figured this article would be worth going through with.

So how does this tantalizing, mouth-watering concept of a perfect home/away rotation pairing work out in the real world of FPL, where danger lurks around every corner? Let’s see.

Aston Villa/West Brom:  By starting out with Brad Guzan and Ben Foster you had a relatively smooth road to GW22; the only hiccup being Foster not playing GW8 where you would have needed to make a decision to use up a transfer to bring in Myhill or to roll with Guzan away to Everton. If you were smart and didn’t use a transfer, this H/H rotation netted you 66 FPL points for your troubles.  HOWEVER, in keeping with my scream of horror on May 2, you would have had to endure 10 of 22 GWs where the away keeper outscored the home keeper.  These fits of blind rage alone would have been enough to send you to the shrink, let alone the fact that if you had went with Guzan only or Foster only you would have bagged 79 and 75 points respectively, thereby completely rendering the H/H rotation strategy unsuccessful at best. Verdict: FAIL.

Burnley/Hull: Investing in Thomas Heaton and Allan McGregor for GW1 would have meant decision time during GWs 7-11 as McGregor did not play. To make matters more complicated, his absence was filled by a combination of Harper and Jakupovic. In hindsight you would have been smart to simply forego a transfer and let Heaton play it out.  After all, he would end up being the 2nd highest scoring keeper for the year, although you wouldn’t have known it at that point. Keeping with the H/H rotation and letting Heaton play both home and away fixtures during McGregor’s absence would have netted you 91 beautiful FPL points, compared to 78 that Heaton earned by himself during that same period and 60 that McGregor earned alone.  Verdict: SUCCESS.

Leicester/Southampton:  Schmeichel and Forster at your service. One a newly promoted keeper and the other a recently signed keeper to a team that was unrecognizable to anyone who watched them dazzle the Premier League the prior year. Lots of question marks here to begin with.  Matters didn’t improve any during GWs 16-22, when Schmeichel was replaced by Ben Hamer. This, the technological equivalent of an energy conserving ARM microchip in a mobile device replacing a far more powerful supercomputer, was cause for concern for the Leicester faithful. Like our other perfect rotations facing the decision of making a transfer or letting the healthy keeper play both home and away fixtures in the injured keeper’s absence, letting Forster endure the Leicester period of uncertainty would have netted you 83 FPL points. Schmeichel all by himself would have earned you 33 points and using Forster exclusively for the first 22 GWs would have earned you 89.  VERDICT: FAIL BUT PRETTY CLOSE

Newcastle/Sunderland:  The rotation we’ve all been waiting for and the one very much in consideration for the 2015/16 season. Krul and who though? This rotation pair didn’t last very long before a big 6’6” wrench in the form of Costel Pantilimon entered the fray. This would have been a tough pair to start with as it is, since there was some uncertainty as to whether Mannone or Pantilimon would get the SUN nod for GW1. As it was, it was Mannone, but only for 9 weeks. Pantilimon then took over and rode the remainder of the season out. This fact, coupled with Krul’s injury problems during GWs 13-20, coupled with the Elliott/Alnwick combination used in his absence, made this a big ole wreck of a rotation pair.  With Mannone only making it 9 weeks and Krul 13,  you were faced with transfer decisions galore, making this rotation pair difficult to decipher.  When all was said and done, if you went Mannone to Pants when it became apparent that he had inherited the first team slot and kept the faith with him during Krul’s absence, you would have saved yourself at least one free transfer and earned yourself a ton of FPL points in the process.  93 FPL points were yours for the taking if you switched Mannone to Pants and then let Pants play home and away through Krul’s injury crisis.  This, compared to the 58 you would have earned with a strictly NEW keeper and the 91 you would have earned by having only Mannone and Pants, although the most roundabout and unpredictable way to get there, results in VERDICT: SUCCESS.

So where does this all leave us?

 Comparing the Rotations to Other Options

In wrapping up this little stroll down Research Lane, let’s see how the rotation pairs noted above compared to other options that were available for FPL managers to consider.  Here, let me make it look pretty for you….

RESULTS/CONCLUSION

So, based on the first half of last season, a few things become apparent.  Interestingly, when looking at value per money spent, 5 of the 8 pairings giving you a return of 9.0 points per million or higher were of the Starter/Backup variety. 2 were Home/Away pairs and only 1 of the 8 pairings above 9.0 PPM involved a premium keeper (DeGea).

If you decide to go with a Home/Away perfectly rotating pair chances are you will have a transfer decision or two to make (or not make) along the way to accompany the brutal pain of watching your away keeper on the bench outplay your home keeper from time to time.  Due to the unpredictable nature of FPL and the increased parity shown across the league, it seems that when one of your rotating keepers goes down injured, letting the healthy keeper roll with both home and away fixtures until you know what’s what with your hurt keeper seems to work pretty well and saves a free transfer to use elsewhere.

Spending more than 9.5m on a pair of keepers definitely seems unnecessary.  Whether you go with a rotating pair of starters (either of the Home/Away or the simple “two cheap starters” variety) or the “starter and his backup” double team option, spending 10.0 to get a top keeper and a 4.0 non-starter to warm your bench seems a bit risky.  Especially when you consider the fact thatthe chances of requiring a transfer to be used to replace an injured or benched premium keeper happened more often than not during the first half of last season. Indeed, spending 10.0 or more on a pair of keepers only would have gotten you 5 of the top 20 scoring pairs, and those were found in slots 5, 8, 12, 16 and 19. Please also note that the points and values for the Hart, Mignolet and Courtois pairs would increase by the amount of points gained by the keeper you replaced them with during their missed fixtures, that is IF you knew they were not going to play and utilized a free transfer to replace them before their missed gameweeks. It is impossible to know what the final results would be for those pairings given the multitude of transfer options available; not to mention the fact that you would have had to burn through a second transfer to get them back once they returned.

If you go with the “starter and his backup” option, an option I find increasingly attractive and one backed up by last year’s stats, you must realize that if a good string of fixtures come down the pike, or a double gameweek is on the horizon, you’ve already doubled down and will only be able to bring in one additional player from that team.  Besides this one potential downfall, you’re guaranteed to have a playing keeper in every fixture (unless of course the issues hitting Hull and Newcastle last year rear their ugly heads, but the likelihood is pretty low – 10% based on last season) and you won’t have to worry about the pain of watching your benched keeper outscore your starter.

When all is said and done, using a perfectly rotating pair of cheap home/away keepers can be effective.  3 of the top 6 scoring pairs in my study based on points were perfectly rotating home/away combinations.  Each of them, however, suffered breaks in the home/away rotation by way of injuries that required the healthy keeper to play home and away fixtures in the injured keeper’s absence.  This is something that you must certainly consider as you get your GW1 planning in gear, since the likelihood of both of your starters being able to play 22 gameweeks of perfect Home and Home rotation to start the season are slim at best.

Either way, it is absolutely imperative that you reconcile yourself with the fact that if you decide to rotate starting keepers, whether you stick to the home keeper every week or pick your starter based on expected fixture ease, you will experience disappointment and see your away keeper outscore your home keeper and your keeper facing a strong side outscore your keeper facing a week side with uncomfortable frequency.  You just gotta get over that shit (talking to myself here).

Due to the huge range of possibilities of having two cheap starters with varying home/away rotation schedules and the subjective nature of trying to pick who to play each week when both sides are either home or away, I did not include those samples in my test.  To do so would not have been practical as every manager will pick his starter based on numerous criteria.  However it deserves to be noted that utilizing this approach and “enjoying” the risk/reward of making weekly decisions based on your own personal criteria certainly could meet or exceed the points and values reflected in this study.  That’s the risk/reward part of FPL that so many adrenaline junkies enjoy anyway, right?

In case you haven’t read the rotation article over at @AllThingsFPL, the perfectly rotating pairs for you to consider (based on expected combined price) in your GW1 planning for 2015/16 include:

Bournemouth/Watford

Crystal Palace/Stoke

Leicester/Swansea

Newcastle/Sunderland

For a quick glance at this year’s budget rotation options plus a glimpse at the premium keeper fixture lists through GW19, the period within which you MUST use the first of your two FPL wildcards, view the table below. Hopefully this will be helpful to you in devising your GK strategy for the first half of the coming season.

Alright; thanks for reading.  If this article helps one person it will have been worth the time it took to research, draft and publish it.  That’s it for my article writing this season…glad to get it over with early 😉

Give me a follow on the Twitter at @FPLWalt and please sign up for our exciting Battle of the Sexes League before the season officially starts tomorrow. The league is having an exciting surge in popularity and you can join by using League Code 6412-8898.

CHEERS,

Walt