Strategic Thinking Cavern (18/19)

[Sutter Kane]

They can't be compared to midfielder prices directly because of the opportunity cost - the fundamental difference between the playing 4-4.5m defenders and the playing 4.5-5m mids. A playing 4-4.5m defender (or indeed any playing defender) automatically has a source of CS points, whereas you need to pay cash for assists and goals for outfielders.

I take your point but I wonder how many points premium defenders have to score before we feel they justify 7-8mn prices (230?)...Our answer to this a couple of years ago might have been 180 imo.

Also if it takes more cash to get a midfielder who'll get you points, just have one less.

Perhaps the value will appear in the midprice midfield bracket, where it was a little absent this time round

Indeed, Fraser (179), Siggy (177), Tielemans will be mid-price next season and if they can get near those scores again will be great value. Also some players who haven't lived up to expectation e.g. Spurs, may be priced a little cheaper. You're right, until we see the prices and the potential bargains, it's difficult to talk concisely about...well, anything.

[Bobby Fetta]

The hypothesis over VAR increasing the penalty count has been proven false across several leagues since its introduction. The world cup was an outlier in how it increased the penalty count, unfortunately, as it is the tournament with the highest profile this falsehood has stuck in the public collective consciousness:

https://www.economist.com/graphic-detai ... alty-count

Thanks, this was really interesting.

But....(being that annoying person who asks for evidence and then ignores it )... I wonder if using the whole season is masking a short-term effect when it is first introduced. I can imagine a couple of weeks of chaos with penalties galore (like the start of the world cup) then defenders work out what they can get away with and it settles back to normal over the season. How about this for a scenario?
- gws 1 and 2 - lots of penalties awarded, penalty takers and a couple of penalty saving GKs at the top of the points tables
- transfer stampede to get penalty takers into FPL teams
- from that point on the pens dry up

On the other hand, reasons for var not to have this dramatic (short-term?) effect:
- var is more established now - players and managers have seen it in action in world cup and european games (even if only tv for some)
- var reduces dodgy penalties awarded for dives

[Bobby Fetta]

I take your point but I wonder how many points premium defenders have to score before we feel they justify 7-8mn prices (230?)...Our answer to this a couple of years ago might have been 180 imo.

I agree with this.

Stem, maybe I am misunderstanding your point regarding base price alternatives but:
- As you said, this year has been odd with a few £4.0m defenders playing, normally you need to spend £4.5m to get a playing defender, same as a midfielder. But we're only talking £0.5m difference anyway.
- Either way, you are really unlikely to get more than 3ppg out of the budget defenders collectively over a season
- A £4.5m mid will probably give 2.5ppg (when needed from the bench to cover rotation etc)
- TAA, Robertson and VVD are all averaging over 5.5ppg this season. You need to spend £10m+ to get a mid or attacker (other than Wilson) getting more than 5.2ppg. Its 1ppg more than nearly all the midprice attackers we spend so much time considering / buying and selling over the season. So they still seem good value, even if they are £8m (I suspect they'll be £7m and £7.5m next season).

When dod wrote words to the effect that we all play this game badly, I really suspect that defender selection is the main area where this applies. Personally I find it indefensible that I haven't owned TAA all season - a £5m player averaging 6ppg. Part of that was keeping a Liverpool spot open in case I wanted to double up in attack (mainly I didn't even do that). I will see if I can stick to this approach next year!

[Stemania]

They can't be compared to midfielder prices directly because of the opportunity cost - the fundamental difference between the playing 4-4.5m defenders and the playing 4.5-5m mids. A playing 4-4.5m defender (or indeed any playing defender) automatically has a source of CS points, whereas you need to pay cash for assists and goals for outfielders.

I take your point but I wonder how many points premium defenders have to score before we feel they justify 7-8mn prices (230?)...Our answer to this a couple of years ago might have been 180 imo.

Also if it takes more cash to get a midfielder who'll get you points, just have one less.

Ha, yeah, for sure there will be a threshold at which it makes sense (though depending on the precise set of player prices next year rather than a fixed number). Certainly anyone who had Mane instead of Robertson all season (like me) can also look with retrospect and think "Bloody hell, I could just have played 433 and had 3m extra to spend", but it's really a question of being able to predict these totals. I think a 200 point season for Mane was feasible before the season, but for Robbo it would have been madness to predict such heights. 200 points for any defender at just about any FPL-realistic price would be attractive next year, but it's a question of knowing/being able to predict what they will get. If we're honest, I think it's fair to say no defender is likely to get over 200 again next season; Liverpool have had a ridiculously good defensive year at a time in a crazy season where the second best team is going to get over 94-97 points. Perhaps we could predict 180 for Robbo/VVD next year (?), and then it becomes interesting. Then you've got to consider that if the best Liv defs are 8m+and the best 5.5m defenders (hard to predict the individuals of course) are probably getting 140+, that 2.5m might still go a long way further elsewhere. This is the beauty of FPL I guess - year on year there is a churn, where the best performing players from one year are hiked and the whole dynamic is changed. We're just at a year where some of the big performers happen to have been defenders.

(Sorry, this is 2/3 of a bottle of wine Stemania talking, with a baby change soon after the word Robbo, so some of it may not be cojent )

Some really nice posts by all in the last day or so incidentally.

[Finisher1]

Having an ambition to pick numerous expensive defs is one thing, but next year the likes of Salah, Mane, Sterling, Aguero, Kane, Auba, Hazard (if still around) etc will all be super-expensive too, so cash might be very tight if the same teams dominate. Perhaps the value will appear in the midprice midfield bracket, where it was a little absent this time round, who knows? That's the fun of it I suppose.

But we will have a super bargain Pukki who will trash all premium forwards and it will have a big impact on our formations.

[Sutter Kane]

Changed than n for an m before my stupid joke.

[Sutter Kane]

Yeah Pukki worth a go if fixtures not awful.

[TheRumourMill]

The hypothesis over VAR increasing the penalty count has been proven false across several leagues since its introduction. The world cup was an outlier in how it increased the penalty count, unfortunately, as it is the tournament with the highest profile this falsehood has stuck in the public collective consciousness:

https://www.economist.com/graphic-detai ... alty-count

Thanks, this was really interesting.

But....(being that annoying person who asks for evidence and then ignores it )... I wonder if using the whole season is masking a short-term effect when it is first introduced. I can imagine a couple of weeks of chaos with penalties galore (like the start of the world cup) then defenders work out what they can get away with and it settles back to normal over the season. How about this for a scenario?
- gws 1 and 2 - lots of penalties awarded, penalty takers and a couple of penalty saving GKs at the top of the points tables
- transfer stampede to get penalty takers into FPL teams
- from that point on the pens dry up

On the other hand, reasons for var not to have this dramatic (short-term?) effect:
- var is more established now - players and managers have seen it in action in world cup and european games (even if only tv for some)
- var reduces dodgy penalties awarded for dives

No problem! Thats a fair point, I have found this article regarding the impact of penalties on Serie A in its opening 5 games of the season:

https://webcache.googleusercontent.com/ ... clnk&gl=uk

To summarise, Penalty kicks given in the opening five Serie A games of the season, in fact, have increased by 63%. An average of 13.5 penalties were given in the opening five rounds of the previous four Serie A campaigns but the introduction of the VAR has skyrocketed the number of penalties allowed in Italy this season to 22 across the opening 5 rounds. So there may well be legs to your theory, Bobby. Small sample size though of course. I'll see if I can find anything similar on other leagues.

[TheRumourMill]

Whilst we're on the subject of price hikes, I'm stunned to see that even Jordan smegging Pickford is on course to outscore Son, Sane, Eriksen, Richarlison, Bernardo Silva, Anderson, Jota, Firmino, Kane, Azpi and Digne. Anyone else getting a 9m Pickford into their GW1 squads? Think I might lobby FPL towers for a play 2 keepers chip!!

[FranckKessie]

They can't be compared to midfielder prices directly because of the opportunity cost - the fundamental difference between the playing 4-4.5m defenders and the playing 4.5-5m mids. A playing 4-4.5m defender (or indeed any playing defender) automatically has a source of CS points, whereas you need to pay cash for assists and goals for outfielders.

I take your point but I wonder how many points premium defenders have to score before we feel they justify 7-8mn prices (230?)...Our answer to this a couple of years ago might have been 180 imo.

Also if it takes more cash to get a midfielder who'll get you points, just have one less.

Ha, yeah, for sure there will be a threshold at which it makes sense (though depending on the precise set of player prices next year rather than a fixed number). Certainly anyone who had Mane instead of Robertson all season (like me) can also look with retrospect and think "Bloody hell, I could just have played 433 and had 3m extra to spend", but it's really a question of being able to predict these totals. I think a 200 point season for Mane was feasible before the season, but for Robbo it would have been madness to predict such heights. 200 points for any defender at just about any FPL-realistic price would be attractive next year, but it's a question of knowing/being able to predict what they will get. If we're honest, I think it's fair to say no defender is likely to get over 200 again next season; Liverpool have had a ridiculously good defensive year at a time in a crazy season where the second best team is going to get over 94-97 points. Perhaps we could predict 180 for Robbo/VVD next year (?), and then it becomes interesting. Then you've got to consider that if the best Liv defs are 8m+and the best 5.5m defenders (hard to predict the individuals of course) are probably getting 140+, that 2.5m might still go a long way further elsewhere. This is the beauty of FPL I guess - year on year there is a churn, where the best performing players from one year are hiked and the whole dynamic is changed. We're just at a year where some of the big performers happen to have been defenders.

(Sorry, this is 2/3 of a bottle of wine Stemania talking, with a baby change soon after the word Robbo, so some of it may not be cojent )

Some really nice posts by all in the last day or so incidentally.

Madness for Robbo to hit 200 is a bit harsh imo Extrapolating his numbers from last season would have given him a 191'ish score. His numbers last year was clearly great, but no real competition for his place, a new world class keeper and I would say that it was also feasible.

I think it would be refreshing if FPL decided to shake things up, because clearly Robbo and TAA justify a price around 9M. The gap within a team has it limits, because of how CS-points work, but equally important of stretching the top of defender prices I think is raising the base price. I'd make defenders have the highest base-price, attackers the lowest and mids in between.

[Stemania]

I'd love them to significantly push up the defender/GK base price - would really shake up the game (and, well, just makes so much sense)!

I guess the issue is that they also don't want to price unnecessarily highly any backup players who likely won't play - and that's often where the fun little gems come from (AWB the main example this year) - I do quite like that aspect too. But, the fact that Wolves' Bennett (who was always likely to start) was priced at 4m was ridiculous, for example. Even if FPL did the prices before the bookies had Wolves down as an unprecedented 8th favourite for the league back in September/August.

[Mo Bot]

I'd like them to sort out the midfielder/striker classification for next season. In keeping Salah, Hazard etc. in the midfield, they have made a rod for their own back with players like Rashford who played half his games as central striker and half out wide. Him and Lukaku swapped places so does that make Lukaku a midfielder?

The striker position is becoming a bit obsolete. Only 6 gameweeks had a striker as top points scorer and two of them were in short gameweeks and another was King v Chelsea which was by no means a banker.

[Stemania]

I guess it's always difficult for them to convert something analogue to something digital. There's the whole DM and wing-back argument too.

Surely it's very possible Salah and/or Mane could get switched to strikers for next year (especially Salah, since he played CF for a bit). Obviously the longstanding classification is that wide forwards in a 433/4231/343 count as mids, and central forwards (or both forwards in a 442/352) count as strikers. But, for Liverpool there's a strong argument to change it I think, because they get their width from their fullbacks with all three forwards playing quite central. With say City, on the other hand (who get their width from there wide forwards), it would be a bit odd for Bilva, Sane, Sterling to be classed as strikers.

Jota will be a striker next year.

[blahblah]

I posted elsewhere about expecting some Mids being Fwds next season: while we may argue etc about which players, I doubt we eould disagree about the sentiment?

Im not digging on this for a few months, but I expect the Citeh and Lpool defenders have pretty much max'ed out on points this season re CS's and possibly attacking points for Lpool.

Did anyone rotate GK's? If so how did it go?

[Turd Ferguson]

Did anyone rotate GK's? If so how did it go?

I intend to write something on GK rotation. In short, it went poorly for me. I rotated Fabianksi and Patricio between my first and second wildcards. When I get the time I'm going to do a deeper analysis into whether it was bad luck or whether it's simply not a viable strategy.

[Turd Ferguson]

Madness for Robbo to hit 200 is a bit harsh imo Extrapolating his numbers from last season would have given him a 191'ish score. His numbers last year was clearly great, but no real competition for his place, a new world class keeper and I would say that it was also feasible.

I agree. I thought Robertson was an obvious pick at the beginning of the season. Liverpool improved noticeably in the second half of last year, and Robertson was flashing attacking threat as well. My back of the napkin calculation had him at 185. Throw in a few clean sheets above expectations for Liverpool, and it gets him to 200. Liv defenders are also helped by the fact that Salah is miserable in the bonus system, and Mane isn't great either. A Liv clean sheet trumps a single goal from Salah or Mane in bps almost every single time.

I do expect them to regress next year. They're not the deepest squad and they caught more than their share of breaks this year judging by xGA. TAA at 6.5 would still be worth every penny though.

One thing to note is that Everton tightened up considerably in the 2nd half of the year. Seems like Silva might have finally gotten them to click. If they price Digne at 5.5, he's prob a season keeper.

[Joccki_10]

Hi all, I’d like to thank everyone who has contributed to this thread during the season. It wasn’t as much as it has been during other years but valuable and friendly nonetheless. It might sound strange but I’m disappointed in how my FPL season has gone because once I broke into the top 1/10k I couldn’t make it any further. (It even felt boring at times!) Normally I’m doing well chipwise but also that has been a disaster this season; haven’t found the reason for that though I can’t stand losing so I blame luck.

Well, I think I’ll be out for a while now. After all I need to find myself a new club. In the GW38 Preview article TopMarx wrote a few things about what has happened to me and there’s a link in there somewhere which directs you to a video of mine for those of you who’d like to have seen me play football.

I’ll be hoping to come back with lots of energy next season to reach a top 1k finish. Thanks again all.

[Stemania]

Seconded, fantastic contributions by all. Special mention to BF and his xG plots.

We may have lulled midseason, but it's a thread for when needs must. This year's preseason was extraordinary useful, and is what I put my decent year down to in the most part. Can't wait till next season when Mav will hopefully return and we can work on the spreadsheets again.

And same to you Joccki. Have a good summer. You better not turn out to be Ashley's Rondon replacement btw!

[Joccki_10]

And same to you Joccki. Have a good summer. You better not turn out to be Ashley's Rondon replacement btw!

And why is that? You’re not satisfied by my video?

[blahblah]

Did anyone rotate GK's? If so how did it go?

I intend to write something on GK rotation. In short, it went poorly for me. I rotated Fabianksi and Patricio between my first and second wildcards. When I get the time I'm going to do a deeper analysis into whether it was bad luck or whether it's simply not a viable strategy.

In time I'll dig out the stuff done last summer, which pointed to it being pointless - hopefully someone can dig out the same data ie GK points home and away plus the different tiers of the Prem....

[MPTree]

Yes, thanks all! I haven't contributed anything this season, but this thread is always a huge help and a key reference point for me throughout the season. It's the people who post here that make it worthwhile.

And congrats to Stem for his extraordinary season, no doubt boosting his ranking in the HoF. We are very lucky to have one of the finest FPL managers in the world share his thoughts so readily with the rest of us. Cheers, Stem!

[Stemania]

Meh, I'm way overrated if that's my rating. One really good season (at last!) compared with, say, Fabio's ridiculous 1.1k, 671, 95th in the last 3 years or Ville's bonkers four top 1k's in six seasons run. I'll soon drop back down in the HoF when the weighting starts to hit.

The odd thing is, looking back it's one of the seasons I feel I remember the least about come the end, having moved house and had a baby somewhere early/mid season in a blur. Annoying really. Normally I'd look back over my transfers from right at the start and seemingly even remember the thought processes involved in those transfers, and could sortof reflect on it for the next year. But, say, I can barely remember owning Trippier and Maddison around GW10 at all, but apparently I did. Maybe that's why it went so well - because mentally there was less of me there making the decisions than usual - only a few Ings-ups this year and a disappointing lack of Phil Joneses.

I suppose if you cling on the the futile hope that cheapo strikers and 5m defenders will save the cash to greedily plug elsewhere without effecting your team every year, then eventually you'll hit the jackpot and fluke a TAA + Jimenez.

[blahblah]

Less thinking is a positive, imho, and something I am going to harness next season. (Although not knowing when the baby was born is pushing it? )

[Sutter Kane]

Picking your battles...

It would be interesting to get a group view on this. I try hardest to play points pure, never looking at ownership %. This can obviously backfire, a captain score of 500 this season proves that! So ignoring the first part of the season which is a little haphazard, do we make sure we cannot be crushed in a GW before even thinking about maximising points? Many solid FPL players would have made sure they have Aguero in GW25 (even seeing people taking hits for Man C players) even if he might only play once - is this relevant that he may only play once? Fact is, he might play twice. There are a few more big decisions throughout the season and I get the feeling I can improve by negating these big swings of points by just going with the 'flow', getting who I need to get and captaining who I need to captain to stay in touch and instead pick the many winnable battles.

[blahblah]

This is interesting, but over a run of GW's as opposed to just the 1, even if it is a DGW.

[Sutter Kane]

Yeah exactly, that too. It's easier over a run of games because the expected points usually edges towards a mean but still, plenty of examples of highly owned played going on unbelievable runs of form. Is it a case of "just get them in ASAP" because of their ownership?

That last sentence feels really alien to me. It's something I've never thought of doing, ever, but I know SpiderMatt was a big believer in it. And he won!

[Sutter Kane]

Oh this brings back memories of Michu...everyone had him and Yaya, and Fellaini. No, not me. They can't keep it up, surely.

[Sutter Kane]

JUST GET THEM IN NO MATTER WHAT!!!

[TheRumourMill]

Picking your battles...

It would be interesting to get a group view on this. I try hardest to play points pure, never looking at ownership %. This can obviously backfire, a captain score of 500 this season proves that! So ignoring the first part of the season which is a little haphazard, do we make sure we cannot be crushed in a GW before even thinking about maximising points? Many solid FPL players would have made sure they have Aguero in GW25 (even seeing people taking hits for Man C players) even if he might only play once - is this relevant that he may only play once? Fact is, he might play twice. There are a few more big decisions throughout the season and I get the feeling I can improve by negating these big swings of points by just going with the 'flow', getting who I need to get and captaining who I need to captain to stay in touch and instead pick the many winnable battles.

On one of the FFS "meet the manager" videos (I think it was Marc Rubenstein?) he made the point that its important to "play not to lose" as often as it is to "play to win". Words that have stuck with me.

[Neath boy]

I never consider % ownership. You have to play the players you believe in.