With the winter Wildcard now available I’ve updated my analysis of the effectiveness of goalkeeper rotations, which I wrote earlier in the season at the Gameweek 9 mark. In this article I review the performance of goalkeepers up to Gameweek 20, particularly as rotation pairs.
OPTIONS
The goalkeepers can be grouped together into four price-bands, based on their starting prices:
- Premium (5.5 upwards): Thibaut Courtois, Joe Hart, Wojciech Szczesny, Tim Howard, Hugo Lloris, David De Gea, Simon Mignolet.
- Mid-price (5.0): Fraser Forster, Adrian, Asimir Begovic, Julian Speroni, Lukasz Fabianski, Alan McGregor.
- Budget (4.5): Costel Pantilimon, Tim Krul, Kasper Schmeichel, Tom Heaton, Rob Green, Brad Guzan, Ben Foster.
- Back-ups (4.0): Robbie Elliot has shown that even owning the back-up keeper to your first choice (Krul) doesn’t guarantee a starting keeper every week.
For rotation combinations I’m focusing on the mid-price and budget options, assuming that few people would consider a rotation strategy involving a premium keeper.
METHOD
I’ve looked at the scores so far for the 20 first-choice goalkeepers. I’ve also calculated the total score for the 63 possible rotation combinations of the mid-price and budget keepers (up to a combined starting price of 9.5). To give an unbiased assessment of which keeper to choose, I’ve based the rotation selection purely on the clean sheet potential of the opposition (using the Fantasy Football Scout season ticker ratings from early in the season). If you make your selection based on form of your keepers it is possible you could get higher scores.
RESULTS
This table shows the highlights and lowlights in the results. The fourth column shows the increase (or decrease) in points compared to the better-scoring single goalkeeper of the combination. Note for the ‘Sunderland’ combination I’ve assumed an initial selection of Vito Mannone for Gameweeks 1-9, with him benched in Gameweek 10 and then transferring to Costel Pantilimon for a four-point hit in Gameweek 11.
The results indicate that:
- A good rotation combination can deliver significantly more points than a single goalkeeper. The option of Fraser Forster plus a cheap non-playing backup is outscored by 24 of the two-keeper combos costing 9.5 or less. Six of these rotation combinations only cost 9.0.
- The best combination so far has been Heaton/Fabianski (109 points), which has delivered 28 points more than Forster alone (81 points).
- The best 9.0 combination has been Krul/Sunderland (104 points). This is still better than any rotation combination involving premium keepers.
- The best single club combination is Sunderland, with the Mannone/Pantilimon double-act delivering 85 points between them.
- Despite their defensive fragility, the predictability of QPR’s performances has made Rob Green a decent option as part of a rotation pairing. He only has four clean sheets but they came in the games you’d expect (SUN, AVL, BUR, CRY).
- Not all rotations have worked well. The Guzan/Foster combo has unexpectedly only earned 63 points, despite perfect home/away rotation. This total is less than Brad Guzan alone (75 points) who is the best single budget keeper so far. Ben Foster has generally disappointed as part of a combination because four of his six clean sheets have come away from home, including difficult matches at Spurs and Southampton. Likewise, two of Asmir Begovic’s four clean sheets were wasted in rotations as they were away at Man City and Everton.
- With only one clean sheet so far and now out injured, many of the worst combinations involve Kasper Schmeichel.
- Tim Howard (41 points) stands out as a disastrous single-choice premium goalkeeper selection, highlighting Everton’s defensive woes this season.
“BUT I ALWAYS END UP BENCHING THE WRONG KEEPER…”
Whenever rotation is discussed someone will always say they don’t like it because they make the wrong selections and leave lots of points on the bench. It’s inevitable that they’ll be unexpected results and disappointments along the way. However, with a good initial choice of keepers and logical weekly team selection, I believe that rotation should work out well over the season. In this graph I’ve plotted total points (horizontal) against points improvement over a single goalkeeper (vertical). A good combination will be above the horizontal line and on the right-hand side of the chart. The graph shows that many of the combinations do perform well. Of the 63 combinations up to 9.5, 44 give an increase in points, whereas 17 give a reduction compared to the better single keeper. The biggest improvement over a single keeper is 34 points, whereas the biggest loss is 13 points, suggesting the potential reward is greater than the risk.
CONCLUSION
I’ve analysed the performance of rotating goalkeepers over the season so far. The data indicates that two cheap keepers offer a viable strategy that can deliver more points than a single expensive keeper. The key is to select keepers from teams that not only achieve clean sheets, but do this in a predicable manner. Saata’s excellent community article on goalkeeper rotation has already identified the best fixture combinations for the budget keepers.
In the results, I’ve shown the combinations that have worked best so far. Remember, however, as with stocks and shares, past performance is not necessarily a guide to future results.
