Sunday, August 3, 2014

Fantasy Football - 2: Picking The Goalkeepers

Edit on 4th Aug - Some of the numbers were slightly off in the rotation analysis. Fixed it. Overall conclusion not impacted though.

Lab Record Type Section


This post is part 2 of a series of number crunching posts I intend on doing on some EPL Fantasy Football data I have. I focus on selecting the goalkeepers in this one.

Part 1 has the main introduction and some other relevant details and assumptions and things all that.

I notice that Lloris' jersey seems to be switched with Begovic's and that Ruddy's jersey is a QPR jersey. This is because the premier league has somewhere update them in such a manner. It is too much pain to manually correct it so  skip it and concentrate on the numbers.

Relevant Summary of Earlier Parts

I typically don't invest a lot in goalkeepers as they only bring extra points through the occasional clean sheets and the rare penalty save. Clean sheets are more of a collective effort than individual brilliance and are unreliable.

The fact that you need two keepers on the rolls to fill one active position is a great opportunity to exploit some bench strength. I typically won't be found with a Cech (Courtouis?) in my team, I'll try and manage by rotating a Mannone and a Speroni maybe.

Szczesny makes it to the top 11 for the year but at 5.5 is just 0.5 too expensive for my liking. His inclusion might be a result of my excluding the bench rotation option.

Weekly Analysis

If you break it down to the week level dream teams, Szczesny with a starting price of 5.5 makes an appearance all the way at rank 6 with 2 appearances in the 37. Ironically, erstwhile number 3 to WS, Vite Mannone, priced at a humble 4.5, playing about 900 minutes lesser than WS, heads the list with 5 appearances.

Name Team Frequency in Weekly Dream Team Start Price Total Minutes Total Points
Vito Mannone SUN 5 4.50 2478 133
Julian Speroni CRY 3 4.50 3330 144
Asmir Begovic STK 3 5.50 2880 136
Ben Foster WBA 3 5.00 2058 90
Adrián San Miguel del Castillo WHU 3 4.50 1710 81
Wojciech Szczesny ARS 2 5.50 3330 157
Joe Hart MCI 2 6.50 2700 111
Tim Krul NEW 2 5.00 3150 121
Artur Boruc SOU 2 4.50 2487 123
Jussi Jääskeläinen WHU 2 5.50 1620 88

If VM had played the same number of minutes as WS at the same levels of performance, he would have reached 179 points. Had VM still been at Emirates while putting in this performance, he would have been bought out by the folks at Etihad. They would have then bought Caballero anyway and Arsenal would still have to settle for Ospina as second choice.

In other news, Simon Mignolet, Bradley Guzan, Allan McGregor, Boaz Myhill, Tim Howard, John Ruddy, Hugo Lloris, David Marshall, David De Gea, and Mark Schwarzer make only one appearances each. Joe Hart has 2 although you may argue that he lost valuable minutes (630 minutes or 7 games) during the butter finger weeks. Notably, Cech does not figure on this list even once.

All the slightly expensive options, top four contestants and whatnot, might be very consistent but from a best team point of view, it (pretty much) never makes sense to make a heavy investment in the keeper; those funds are better used elsewhere. The top 5 on the above list all played for teams which were near the relegation zone at some part of the season. Moreover, West Ham had their minutes split between two keepers, and Boruc was injured otherwise Szczesny might have been even lower than number 6. The defenses at the lower end of the table are going to play their hearts out during various stages of the season and are worth keeping on the payroll.

The Value of Rotation

The goalkeeping position is that it is the least complicated selection in the team. Only one keeper plays at any time and there are two available to pick from. This is a very important factor in how I pick the keeper because it allows me the freedom to pick two low-priced keepers who can complement each other whenever possible and still fetch a decent tally with rotation. Mannone and Speroni for me any day over Cech and Hart.

If we had two keepers whom we rotated such that we always picked the higher scoring keeper for that gameweek, then we could have upped our goalkeeping returns significantly as the Total Points column in the below table indicates.

RankName 1Team 1Start Price 1Name 2Team 2Start Price 2Total Start PriceTotal Points
1 Tim Howard EVE 5.50 Vito Mannone SUN 4.50 10.00 226
2 Asmir Begovic STK 5.50 Vito Mannone SUN 4.50 10.00 223
3 Vito Mannone SUN 4.50 Hugo Lloris TOT 6.00 10.50 221
4 Wojciech Szczesny ARS 5.50 Tim Howard EVE 5.50 11.00 217
5 Petr Cech CHE 6.50 Tim Howard EVE 5.50 12.00 217
6 Wojciech Szczesny ARS 5.50 Vito Mannone SUN 4.50 10.00 215
7 Wojciech Szczesny ARS 5.50 Petr Cech CHE 6.50 12.00 214
8 Wojciech Szczesny ARS 5.50 Julian Speroni CRY 4.50 10.00 213
9 Julian Speroni CRY 4.50 Artur Boruc SOU 4.50 9.00 212
10 Petr Cech CHE 6.50 Vito Mannone SUN 4.50 11.00 212
11 Julian Speroni CRY 4.50 Hugo Lloris TOT 6.00 10.50 211
12 Artur Boruc SOU 4.50 Vito Mannone SUN 4.50 9.00 208
13 Wojciech Szczesny ARS 5.50 David De Gea MUN 6.00 11.50 206
14 Tim Howard EVE 5.50 Artur Boruc SOU 4.50 10.00 206
15 Wojciech Szczesny ARS 5.50 Asmir Begovic STK 5.50 11.00 206
16 Julian Speroni CRY 4.50 Tim Howard EVE 5.50 10.00 205
17 Tim Howard EVE 5.50 Simon Mignolet LIV 5.50 11.00 205
18 Tim Howard EVE 5.50 John Ruddy NOR 5.00 10.50 205
19 David Marshall CAR 4.50 Vito Mannone SUN 4.50 9.00 205
20 Tim Howard EVE 5.50 Hugo Lloris TOT 6.00 11.50 204
Team of the Year member, Szczesny costs 5.5m and brought in 157 points over the seasnn. We had also set aside 4.5m per bench member so you could say that we have to pick two keepers within 10m. The top pair of Howard and Mannone costs 10m and could have brought in up to 226 points during the year. If we did really, really want to keep Szczesny then his best pairing, (surprise, surprise!), would also be with Mannone. With that choice, we would lose 11 points to reach a maximum of 215 which is still 58 points more than what Szczesny earned alone. If we were willing to sacrifice 3 more points to come to maximum of 212, we could instead select Boruc and Speroni and save an extra 1m in that process. Speroni by himself secured 144 points over the season which is only 13 less than Szczesny's 157.

If you look at last year's fixture list for Boruc and Speroni, in 31 of the 37 games at least one of them had a home fixture. If you were to just pick the home game keeper, the average of the two if both had home games, pick Speroni for the game Boruc was injured or didn't play, and use your intuition to pick the keeper for the three games left after this filtering, you could still manage 156 which is just 1 point lesser than Szczesny by himself. During the actual game itself, you might not blindly choose home game over away game when say City during their score-8-in-each-match phase were playing the keeper at home while West Ham during their Carlton-Cole-upfront phase were playing the keeper away but the point is that a very simple rule based selection could keep the points tally up but also save money.

I wanted to see if there is a correlation between this complementing fixture set and the pairs that figure above but because of mismatching double gameweeks and absence in other gameweeks, it is hard to get a correct number. I manually checked it for Howard and Mannone and their fixture list complement each others too. A similar rule as the one above, gets you 180 points from this pair, 23 points more than Szczesny. There's probably something in it.

This investment of 1m in the outfield could increase the Team of the Year tally by 9 points from 2110 to 2119 by replacing Mertesacker and Lambert with Fonte and Sturridge. The net gain would be of ((156 - 157) + 9 =) 8 points. That doesn't seem much but remember that we are taking the simplified scenario without bench rotations and transfers. An extra 1m would be quite handy in those matters.

However, keeping that extra 1m in the goalkeeping budget and investing it in Howard + Mannone instead of Boruc + Speroni would increase the tally by 23 points and maybe even more if Mannone had been playing from the start. Think Mignolet, or Begovic during the season before that. Foster, or Krul in 2011-12. There will always be someone.


Unless you are as risk averse as someone who keeps all their savings in FDs, avoid the Cechs, and the Courtouiss and the Harts. Stick to a good mid + low-budget keeper combination and rotate them wisely. A keeper with idiot defenders gets to make a lot of saves and gets you extra points for the saves, if not for clean sheets.

When picking out keepers, track the manager's strategies (Allardyce, Pulis, Poyet, etc.) and see when they switch to survival mode and enforce a defensive clampdown. Also check out the fixture list to see possible complementary home/away fixtures. On second thoughts, let me do it myself - here, you're welcome.

My prediction for the top 20 rewarding combinations are the ones below. No City or Chelsea. Plenty of bottom-of-the-table teams. No LIV in case Reina and Mignolet get rotated, if they add to their defense and decide on one keeper then maybe rethink.

Team 1 Team 2 Complementary Home Games

Other Posts in the Series

Part 1: General Team Composition
> Part 2: Picking the Goalkeeper
Post 3: Picking the Defensive Line
Part 4: The Value of Money
Part 5: The Value of Wildcards
Part 6:  Getting Ready for 2014-15
Part 7: How Much Does the Opposition Matter?

No comments:

Post a Comment