To see less ads Register or Login ----- Daily Fantasy Sports games 18+

Dod's Blog

A forum for comment and discussion on Fantasy PremierLeague.com (FPL) Teams. Post your Rate My Team (RMT) messages here!
Post Reply
User avatar
blahblah
FISO Viscount
Posts: 108502
Joined: 13 Oct 2005, 18:46
Location: .. he thinks that he knows something which he doesn't, whereas I am quite concious of my ignorance.

Re: Dod's Blog

Post by blahblah »

That example is far from extreme...

The general idea seems very similar to how I used to play Private Leagues (when I was quite good 😂): cover the obvious and gain a lead through differentials then reduce the risky differentials; if behind then reassess which players to cover\differential.

So in the FH example if both could see the other team then they may both willingly settle for 1, 2 or 3 different players or just the Capt choice....

User avatar
Ironfist
FISO Knight
Posts: 13150
Joined: 11 Jul 2009, 14:43
Location: Israel
FS Record: FPL 09/10 - 1824th
Contact:

Re: Dod's Blog

Post by Ironfist »

An entertaining Tolkienesque saga cruelly crashing into reality with Stemania hacking into your account... scary :twisted:

User avatar
Patrician
Grumpy Old Man
Posts: 1492
Joined: 13 Oct 2005, 18:32
Location: ...bears no responsibility for bad points forecasts
FS Record: 2nd TFFE 2008, FPL 3 top 10k finishes
Contact:

Re: Dod's Blog

Post by Patrician »

dod wrote: 09 Jul 2019, 20:38 I have many of the same reservations as Stemania regarding the value (or otherwise) of setting a points target to pursue. My main problem is this: I'm not sure I want to convince you that you are wrong :lol: . You are generating so much interesting and useful by-product from your pursuit of this hypothesis that it really doesn't matter if I personally disagree with your reasoning for doing so.

What if the reason that NASA went to the moon was to harvest the green cheese? We'd still have freeze dried food, memory foam, and solar cells, even if not a single gram of green cheese was discovered :wink: .
Aha, light dawns on Marblehead. The innovation is the point :wink:

Perhaps if I update my FPL goals to be "I aim to win FPL with a world record points total for any season".

User avatar
Smurphy Paw
FISO Knight
Posts: 14504
Joined: 17 Aug 2006, 17:48
FS Record: Mediocre, apparently
13/14: FPL 1792; FIFA 14 Top 700.
17/18: FPL 696th; loads of mini-League wins and side game promotions
18/19 1FC Köln 5AS Champions
#1 Spring Super League regular season 19/20 & 20/21

Re: Dod's Blog

Post by Smurphy Paw »

blahblah wrote: 10 Jul 2019, 06:32 That example is far from extreme...

The general idea seems very similar to how I used to play Private Leagues (when I was quite good 😂): cover the obvious and gain a lead through differentials then reduce the risky differentials; if behind then reassess which players to cover\differential.

So in the FH example if both could see the other team then they may both willingly settle for 1, 2 or 3 different players or just the Capt choice....
FFFix even does this for you. Their ‘Spy’ function predicts the one or two differential transfers to improve your chances against specific opponents

User avatar
Jameselaprendi
Grumpy Old Man
Posts: 2099
Joined: 19 Jul 2006, 23:13
Location: Morons, your bus is leaving!

Re: Dod's Blog

Post by Jameselaprendi »

@Dod

In your responses and I sense you are showing a strong tendency towards a ‘game theory’ perspective. Isolate an opponent, out-manoeuvre them, etc. For that reason I think the main point I was trying to make bounced off, but that could also be because I the way I made it was scattergun and half baked :lol:

In order to manage the risk in such a way that we gain an advantage (/ maximise points :P) we need to take advantage of opportunities of a certain risk/reward profile. Perhaps comparable to playing tight-aggressive in poker - but to be clear I only use that example to evoke a richer sense of aiming to identify opportunities that are lower-risk with higher-upside.

Identifying the opportunities is always a combination of instinct and math and game theory - you ground yourself in the math and game theory and then you let instinct drive. And give it a licence to break all the rules, but erring towards a ‘nuclear codes’ approach to doing so - in order to reign it in. Hope that makes sense 8-) :shock:

I think (hope) what I’ve said so far is too general and somewhat too obvious to disagree with...?

So my question is - why would you arbitrarily play safe at the beginning of the season?

Unlike poker - you can (mostly) only accumulate points - so what have you got to lose? You could only say you could lose ground against the rest of the field (or the one true rival) but I think you can cover that with circa 75% of your player choices - at any given point in the season?

User avatar
blahblah
FISO Viscount
Posts: 108502
Joined: 13 Oct 2005, 18:46
Location: .. he thinks that he knows something which he doesn't, whereas I am quite concious of my ignorance.

Re: Dod's Blog

Post by blahblah »

Smurphy's Paw wrote: 10 Jul 2019, 08:23
blahblah wrote: 10 Jul 2019, 06:32 That example is far from extreme...

The general idea seems very similar to how I used to play Private Leagues (when I was quite good 😂): cover the obvious and gain a lead through differentials then reduce the risky differentials; if behind then reassess which players to cover\differential.

So in the FH example if both could see the other team then they may both willingly settle for 1, 2 or 3 different players or just the Capt choice....
FFFix even does this for you. Their ‘Spy’ function predicts the one or two differential transfers to improve your chances against specific opponents
Someone should try this vs the Leader 😎

User avatar
blahblah
FISO Viscount
Posts: 108502
Joined: 13 Oct 2005, 18:46
Location: .. he thinks that he knows something which he doesn't, whereas I am quite concious of my ignorance.

Re: Dod's Blog

Post by blahblah »

Jameselaprendi wrote: 10 Jul 2019, 08:45

So my question is - why would you arbitrarily play safe at the beginning of the season?

Unlike poker - you can (mostly) only accumulate points - so what have you got to lose? You could only say you could lose ground against the rest of the field (or the one true rival) but I think you can cover that with circa 75% of your player choices - at any given point in the season?
You obviously haven't had my starts 😂😂😂

User avatar
dod
FISOhead
Posts: 734
Joined: 31 Jul 2015, 23:49
FS Record: I have never won FPL.

Re: Dod's Blog

Post by dod »

Jameselaprendi wrote: 10 Jul 2019, 08:45 So my question is - why would you arbitrarily play safe at the beginning of the season?
It's a good question but I think I've already covered it in my guide.

In short:
1/ The start of the season is when you have the least amount of information on which to make an informed decision.
2/ You are disproportionately punished for getting a decision wrong because the market is at it's most active.
3/ A 0.1m drop in a player's value is effectively a 0.5m drop for the first few GWs.

User avatar
Smurphy Paw
FISO Knight
Posts: 14504
Joined: 17 Aug 2006, 17:48
FS Record: Mediocre, apparently
13/14: FPL 1792; FIFA 14 Top 700.
17/18: FPL 696th; loads of mini-League wins and side game promotions
18/19 1FC Köln 5AS Champions
#1 Spring Super League regular season 19/20 & 20/21

Re: Dod's Blog

Post by Smurphy Paw »

dod wrote: 09 Jul 2019, 11:56
Smurphy's Paw wrote: 09 Jul 2019, 07:54 But it always reminds me that history remembers the person for whom that volatile decision went well and fails to remember the rest.
I like that a lot :) . It's a good lesson that I think is applicable to FPL. It may be due to skill or it may be due to dumb luck but getting to the apex usually (always?) involves being on the right side of a volatile decision(s).

I've used Tour de France analogies in this thread myself but I'm not sure that yours quite works. In FPL we all have access to exactly the same tools. The same players, the same chips, and for the most part the same ability to access information that may help us make our decisions. In the Tour de France every competitor has different physical attributes, they are riding different machinery, and they belong to different teams with varying levels of support. Not every rider can sprint like Cavendish or climb like Virenque. Nor are they all riding Pinarello Dogmas, or have Froome as a domestique.
I actually agree that the TdF example doesn’t quite work to make the point you make but that doesn’t necessarily invalidate it. As soon as the GW1 deadline has passed we no longer have equal access to the same players. For example, last season I got locked out of my account 20 minutes before the deadline and had Schneiderlin as a place holder instead of Richarlison who I was trying to bring back in, plus four or five other differentials. It cost me 40 points in GW1 alone; and from thereon-in I didn’t have the optimal Wiggo-Froome set up so had to play differently.
In truth it’s a side alley of a conversation wrt the wider discussion and I almost didn’t post it.
However based on this painful experience I will also be playing more safely with my initial team (& the clock!!), which I suspect aligns me with you in the discussion you’re having with james
I also identified my one true opponent very early on last season. It was myself. And isn’t that what we’re all doing? Seeking to be one point better/one transfer fewer than our ideal self. We only ascribe it to a third person later in the season

User avatar
Smurphy Paw
FISO Knight
Posts: 14504
Joined: 17 Aug 2006, 17:48
FS Record: Mediocre, apparently
13/14: FPL 1792; FIFA 14 Top 700.
17/18: FPL 696th; loads of mini-League wins and side game promotions
18/19 1FC Köln 5AS Champions
#1 Spring Super League regular season 19/20 & 20/21

Re: Dod's Blog

Post by Smurphy Paw »

blahblah wrote: 10 Jul 2019, 09:07
Smurphy's Paw wrote: 10 Jul 2019, 08:23
blahblah wrote: 10 Jul 2019, 06:32 That example is far from extreme...

The general idea seems very similar to how I used to play Private Leagues (when I was quite good 😂): cover the obvious and gain a lead through differentials then reduce the risky differentials; if behind then reassess which players to cover\differential.

So in the FH example if both could see the other team then they may both willingly settle for 1, 2 or 3 different players or just the Capt choice....
FFFix even does this for you. Their ‘Spy’ function predicts the one or two differential transfers to improve your chances against specific opponents
Someone should try this vs the Leader 😎
Just be prepared for a Red Dwarf Holly-esque, ‘If I was heading there, I wouldn’t start from here’!

User avatar
blahblah
FISO Viscount
Posts: 108502
Joined: 13 Oct 2005, 18:46
Location: .. he thinks that he knows something which he doesn't, whereas I am quite concious of my ignorance.

Re: Dod's Blog

Post by blahblah »

4, you won't win it in the first x GW's but you can certainly lose it.

User avatar
blahblah
FISO Viscount
Posts: 108502
Joined: 13 Oct 2005, 18:46
Location: .. he thinks that he knows something which he doesn't, whereas I am quite concious of my ignorance.

Re: Dod's Blog

Post by blahblah »

Smurphy's Paw wrote: 10 Jul 2019, 10:50
blahblah wrote: 10 Jul 2019, 09:07
Smurphy's Paw wrote: 10 Jul 2019, 08:23
blahblah wrote: 10 Jul 2019, 06:32 That example is far from extreme...

The general idea seems very similar to how I used to play Private Leagues (when I was quite good 😂): cover the obvious and gain a lead through differentials then reduce the risky differentials; if behind then reassess which players to cover\differential.

So in the FH example if both could see the other team then they may both willingly settle for 1, 2 or 3 different players or just the Capt choice....
FFFix even does this for you. Their ‘Spy’ function predicts the one or two differential transfers to improve your chances against specific opponents
Someone should try this vs the Leader 😎
Just be prepared for a Red Dwarf Holly-esque, ‘If I was heading there, I wouldn’t start from here’!
Not having watched this and GoT is making feel aloof and superior 😱

But I don't trust Computer suggestions eg if Nor, AV or SU have a run of CS's then imagine the suggestions 😂😂😂 If people think this is mad a few seasons ago WBA kept 4 CS's in the first 6ish GW's and then 2 in the next 32....

User avatar
Jameselaprendi
Grumpy Old Man
Posts: 2099
Joined: 19 Jul 2006, 23:13
Location: Morons, your bus is leaving!

Re: Dod's Blog

Post by Jameselaprendi »

dod wrote: 10 Jul 2019, 10:34
Jameselaprendi wrote: 10 Jul 2019, 08:45 So my question is - why would you arbitrarily play safe at the beginning of the season?
It's a good question but I think I've already covered it in my guide.

In short:
1/ The start of the season is when you have the least amount of information on which to make an informed decision.
2/ You are disproportionately punished for getting a decision wrong because the market is at it's most active.
3/ A 0.1m drop in a player's value is effectively a 0.5m drop for the first few GWs.
On 2 and 3 I should weigh that up as a part of the risk. So I don’t disagree. Maybe just with your interpretation of who that rules out.

If it’s more 1 then I guess it’s just about what information you’re looking for...

In the end I don’t disagree with your principles and it’s shades of grey whether I disagree with your explicit interpretation of these principles. Hope you don’t miss an opportunity early doors!

User avatar
Jameselaprendi
Grumpy Old Man
Posts: 2099
Joined: 19 Jul 2006, 23:13
Location: Morons, your bus is leaving!

Re: Dod's Blog

Post by Jameselaprendi »

Smurphy's Paw wrote: 10 Jul 2019, 10:44 As soon as the GW1 deadline has passed we no longer have equal access to the same players.
Brilliant comment - now it’s something like Risk after you’ve picked your starting locations... Or I believe Magic where you assemble a deck...

Picking the XV is a very different kettle of fish to managing the squad from the moment the clock goes red! That should be a major consideration - “what are giving future you to work with...”

User avatar
Stemania
FISO Jedi Knight
Posts: 20448
Joined: 27 Aug 2006, 11:54
Location: On the Iron Throne of xG, the seat of The Crown Prince of the Stat Perverts. Or if not, in the STC!
FS Record: Best: TFF 321st. FPL 129th. FFS Career HoF Peak 2nd (Live 1st). Ability since lost.

Re: Dod's Blog

Post by Stemania »

dod wrote: 10 Jul 2019, 05:31 I like to think of that 'one rival' in terms of a probability cloud and an incrementally collapsing wave function but really it's just a different way of visualising what you need to do in order to achieve your FPL goals.
Cool, no qualms. I agree we are mostly in agreement. :D
dod wrote: 10 Jul 2019, 05:31 There is one apparent area of difference though. If I understand what you are saying correctly I can agree that maximising xPoints is the best method to achieve the highest total points score but that is not the same as maximising your chances of achieving your FPL goal.
So, with the specific example you give, I agree. I've no issue that a legitimate strategy in the final week or two is to "gamble and pick a team with lower expected points hoping to get lucky on the day" as you put it (though it's arguable how often a player will actually find themselves falling short of, yet still within a suitable distance of, their 'goal' come the end of the season). But, I'd simply point out that the extreme end-cases should be somewhat irrelevant in defining what everyday FPL strategy should be. For gameweeks 1-36, say, this would not apply, and frequently going with lower expected points players is far more likely to put you out of reach of the above last gasp Hail Mary than to help in achieving the season goal. :)

I guess I should flesh out my rather weak 'would not apply' claim. But first, there's an elephant in the room related to something you touched upon. We don't have perfect xPoints creating machines so how on earth do we 'maximise xPoints'? That's a genuine problem with just about everything I ever say. :lol:

My answer is that we all construct our own expected points models, be it in our head or on paper/computer. Some might create physical systems (such as FFS's, or Patrician's points projections), some might base their estimates on the relative stats of the picks being considered (xG, xA etc), some might rank players mostly on the eye test. Nowadays I think it's fair to say most of us use a bit of a mix - I certainly use a bit of all three. Further, each system is usually moderated by another; e.g., if we see a points projection from, say, FFS, then (consciously or not) it goes through a moderation process in our heads to the tune of "well, I think player X's figure is actually a little high due to such and such a factor, and player Y's is low due to such and such a factor". There's so much subjectiveness, that it's nigh on impossible to produce exact figures and have huge confidence in their efficacy. So perhaps it's best to imagine xPoints as a sort of small range of rough points potential, in which the true unknowable value somewhere lies. In that context, it's hard to say what is meant by a "close call". I guess this is what Patrician was getting at. Here, I think there is licence for individual freedom - one person's 'close' might not be another persons 'clear gap', but it's something almost impossible to define methodically unless you produce your own final 'moderated' points projections and you systematically base all decisions on those explicit figures. We all have our own way of deciding such things, and consistency of our own decision process can no doubt be a huge factor in success or failure. :)

On a related note, here is another point I'd like to broach. As mentioned before, when the call is close during the season, we both happily agree that you can do what you want with variance because in theory it makes very little difference to your expected points total. But, here's an argument against that based on the fuzziness of knowledge. Whilst in theory I say it doesn't matter when it is close, in practice it's my opinion that the low-varience option is actually more likely to be the better choice - precisely because the realistic (lack of) accuracy of our xPoints estimates and our fuzzy definition of what is and what isn't a close call based on those predictions. To set the scene, first let's distinguish between types of 'low-volatility' player - there is a category of player who is simply well owned (WO) among the managers against which we are competing, but there is also a (not necessarily disjoint) category of 'low-volatility' player who represent those being heavily transferred in by those managers right now - say, the 'bandwagon' (BW) players - those in the second category essentially have high ownership, it's just not visible in the raw ownership figures yet.

Now, let's say we find ourselves in a situation where we've used all the faculties available to us to decide which of two players is a better pick, and we've come out with "don't really know, it's too close a call". Whilst I remain fairly ambivalent on the WO player (and in that case I see ownership as essentially an irrelevance), I think there there is a strong argument to go with the BW player:
  • 1) We should be humble about our own personal ability to accurately judge players and to produce accurate points predictions. The visible community of FPL managers is very knowledgeable, they're not just a mass of unthinking randomers. We personally cannot see every game, we cannot see every stat, we are not all experts in every aspect of football. Let's say we personally see picking player X as a very close call vs picking player Y (so X=Y), and the community is on mass going with player X (so viewing X>Y). Then, if we were to assign a probability to 'being correct', then it is far more likely that we are slightly wrong and the community is right (X>Y), than we are slightly wrong and the community is even more wrong (X<Y); precisely because the managers against which we are competing are very good practitioners of FPL. Of course, we should base our decisions on the arguments made in support of each player, rather than raw numbers behind each viewpoint, but those arguments should already have been included in the decision making process that lead us to the 'close call' conclusion.

    2) A player who is being transferred in heavily is likely to come with price increase(s), which has potential benefits later in the season.
In fact, some may argue that point (2) is actually enough on its own to sometimes widen the definition of close call and go with a slightly 'worse' xPoints pick on occasion. That's not really my bag in general, but I can certainly see the logic in the argument and have surely gone down that route once or twice - especially if you to attach to every 0.1m gained a certain xPoints value for the rest of the season. :D

Finally, less of a theoretical issue with volatility hunting, but more of a practical issue from my observations. One problem I have with the practice of actively seeking to increase variance is that there is a danger to let the search for volatility drive the judgment of points potential. Basically, as an example it can all be encapsulated in the term 'differential' itself. It comes with a warm fuzzy glow; I might feel I'm 'controlling my own destiny' and feel 'above' the crowd. Such is the want to be different, to be unique, that the fact that a player is not well owned is so often actively used as a key driver in 'why' they are picked, not as a tie-breaker as we've been discussing. Such is the keenness for volatility, that 'differential status' encroaches on what should be a dispassionate and systematic player evaluation process. :)

That brings me neatly back to why I don't think the 'Hail Mary' example you provided can be extrapolated to general play during weeks 1-36, or rather, where I think the idea of increasing variance at the cost of some xPoints is often missaplied early. Law of large numbers aside, our ability to actually judge the makeup of teams above us and discuss this in the community comes across to me as highly flawed (due in no small part to psychological misinterpretation of the available data). As an illustrative example, with a few weeks to go at the end of the last season, the midfield ownership percentages in the top 10k were:

Son (76.9%), Mane (72.7%), Sterling (66.9%), Jota (52.2%), Salah (39.6%)

Ignoring the obvious Liverpool defender double up argument for the sake of example, let's assume we believe those are the 'best' top scoring midfielders to have in our 352/451 formation. Then, so often those picks would be described as 'template' and attempts would be made in the community to actively avoid some of them in order to 'gain ground' on the top 10k. "I can't pick player Z if I want to gain ground because everyone has him". But how many teams in the top 10k actually own 3 or more of them, 4 or more of them, or all five? It will be much much lower than the naiive impression we get from just reading off the figures. It won't be the same 70-odd percent of managers that own Son + Mane; it is absolutely nowhere near 39% or so that own all five. If you were to own all five in that situation you would in fact be beating the vast majority of teams in the top 10k (in terms of your xPoints calcs)! Thus, I see it as very difficult for the 'Hail Mary' GW38 example to be extrapolated to generic play, or even to identify when it would ever be viable outside an extremely small set of situations. The teams above us simply do not all look the same, or anywhere near as similar as that is imagined.

Perhaps when discussing captaincy, actively increasing variance is better considered?

Anyways, good god that was longer than intended, sorry. :mrgreen:

User avatar
Taff Murray
Grumpy Old Man
Posts: 1584
Joined: 23 Jul 2011, 17:19
Location: No.1 At the end of the bar
FS Record: FISO 5AS FA Cup Winner 2016/17, FISO 5AS Champions League Winner 2018/19 Best FPL Pos 11/12 - 9959th

Re: Dod's Blog

Post by Taff Murray »

OMG I wish I understood a single word of that. :shock: :wink: :lol:

Joking aside. Brilliant post. Well thought out and written, if not succinctly, certainly very eloquently.

User avatar
raoul
Dumbledore
Posts: 5031
Joined: 13 Oct 2005, 18:32
FS Record: Tied for OR 1st worldwide in FPL in GW5 in 21/22 (only to then finish 1m), and ranked 1st in TFF late in season 21/22 (eventual finish 95th). Won an old game called Football Fantastic. Best Fanteam finish 117th and cashed £150. Best FPL finish 2.4K.

Re: Dod's Blog

Post by raoul »

Some very useful points worth drawing out here.

- very few will have all the well owned players. Having them all is an advantage over almost all other players, from a probability perspective

- it is too easy to want a player because they have 2% ownership rather than because they seem a good pick for other reasons

- especially once the season is established, the top 10k are mostly there on good rather than lucky choices. If they own a player a lot, that suggests it is a good pick

Btw I must say I find current ownership stats odd. Liverpool defenders in particular. Early days I suppose but...

User avatar
dod
FISOhead
Posts: 734
Joined: 31 Jul 2015, 23:49
FS Record: I have never won FPL.

Re: Dod's Blog

Post by dod »

Patrician wrote: 10 Jul 2019, 07:25 Aha, light dawns on Marblehead.
Many moons ago I had a work colleague - who originated from Boston - who regularly used this expression (although he left out the 'Aha' bit). At the time none of us knew what he meant and assumed it was some quaint Americanism. I was reminded by your post to finally Google it. Turns out that it means what is known in psychology as an 'Aha experience' :lol: .

User avatar
dod
FISOhead
Posts: 734
Joined: 31 Jul 2015, 23:49
FS Record: I have never won FPL.

Re: Dod's Blog

Post by dod »

@ Stemania

It looks like our areas of disagreement are minor. In fact your conclusions pretty much concur with the TL;DR bit at the end of my last theory post :lol: .

There are however one or two points in your long post that I think are very interesting and merit further discussion. I just don't have the time at the moment but will get back to them as soon as I do :) .

User avatar
dod
FISOhead
Posts: 734
Joined: 31 Jul 2015, 23:49
FS Record: I have never won FPL.

Re: Dod's Blog

Post by dod »

What can shogi teach us about how to play FPL?

Some of you may be looking at the title of this post and thinking "Who, or what, is shogi?" :?

If you guessed that it is a moderately powerful Pokemon, a hellspawned Elder God from the fevered imagination of H.P.Lovecraft, or a skin disease caused by poor personal hygiene, then sadly, I am afraid you are wrong. Shogi is in fact the Game of Generals, also known as Japanese chess.

Being Japanese, it is of course far more complex and difficult to play than normal chess :roll: . Think of it as 'chess for clever people.' I know this because some thirty odd years ago I attempted to learn how to play it before quickly deciding that like unicycling, macrame, and eye surgery, it was a skill where the effort to acquire it far outweighed the need to do so.

I did however glean enough about the game to understand what made it so devilishly complex. Unlike in normal chess where a captured piece is permanently removed from the game; in shogi, a captured piece becomes the property of the player who captured it and can be 'dropped' (yes that is in fact the correct term :lol: ) into the game at any point it is that player's turn to move. Because a captured piece could suddenly appear anywhere on the board it was far more powerful than a normal piece of the same type.

It is this concept that I would like to appropriate for FPL. The idea that a resource can be much more powerful when you don't employ it :o . This idea would hardly be news to any student of military history (which I'm definitely not :lol: ). Would Sun Tzu, or Alexander the Great, or Napoleon, have committed all their troops to the battlefield straight away, or would they have kept some in reserve? I don't know ( :oops: ) but I'm presuming the answer is that that they would have kept some in reserve, or if I am wrong that you won't bother to check because it's just common sense innit? :P

As far as FPL goes I am specifically thinking of not spending our entire 100m starting budget on our initial squad but rather keeping some in the bank. I'm only talking about 0.5m here. Obviously you want most of your budget in play earning you points, but is that last 0.5m really going to hurt you that much?

Well that depends. What do you get for this 0.5m if you don't put it into play? There are two main ways it helps you:

1/ It expands the range of players you can choose to replace any player in your squad. If, for example, Sanchez plays as a FWD and scores a brace for MUN in GW1 you will very likely want to consider bringing him into your squad. If you own a 6.5m MID that you would like to turn into Sanchez it's easy. One transfer and you have the spare cash ITB. Without that 0.5m ITB - and unless you specifically own a 7m MID - you will either have to downgrade a higher priced MID or take a hit and make an enabling transfer to raise the necessary funds. Having that 0.5m ITB massively expands your options to jump on early bandwagons with a single transfer for every player in your squad.

This is particularly important at the start of the season with everybody priced in nice round 0.5m jumps and the player price market is at its most active and volatile. It doesn't help you to wait another GW and use 2FTs to bring in your bandwagon player because a) You will still be making an enabling transfer and b) He will likely have risen in price and you will have to find what is effectively another 0.5m in order to bring him in.

2/ The alternative use of the 0.5m ITB is to be able to bring in players who started at the same position and price as a player you own but have risen in price. You won't be able to jump on all the early bandwagons unless you take hits. You also might not want to. Not all that glitters is gold, after all. 0.5m ITB allows you to wait for more information. You are not forced to be the first person to find out if the pretty red berries are edible. You can wait a GW or two to see if anyone dies or spends the next morning with their head in the toilet.

This is a bigger deal than it sounds. Remember, if you jump on a bad bandwagon it will cost you not only the transfer you used to bring that player in but also the transfer you will need to remove them again. That spare 0.5m allows you to be late to the party up to 5 times and each time you correctly change your mind because a player turns out to be fool's gold not only have you saved the pts you would have lost by your rash decision based on inadequate information but also two transfers 8-) .

Whatever else may change between now and the start of the FPL season, one thing I am certain of. My budget will be 99.5m :wink: .

User avatar
Pirlo's Beard
FISO Jedi Knight
Posts: 20554
Joined: 21 Aug 2013, 17:48

Re: Dod's Blog

Post by Pirlo's Beard »

dod wrote: 16 Jul 2019, 00:08 You are not forced to be the first person to find out if the pretty red berries are edible.
No one wants to be the Chris McCandless of FPL. Or of anything, really.

User avatar
dod
FISOhead
Posts: 734
Joined: 31 Jul 2015, 23:49
FS Record: I have never won FPL.

Re: Dod's Blog

Post by dod »

Pirlo's Beard wrote: 16 Jul 2019, 00:35
dod wrote: 16 Jul 2019, 00:08 You are not forced to be the first person to find out if the pretty red berries are edible.
No one wants to be the Chris McCandless of FPL. Or of anything, really.
I had to Google that. I haven't watched the film or read the Jon Krakauer book but I might do in future as I really enjoyed 'Under the Banner of Heaven' by the same author. I have heard of lathyrism and ODAP before though.

User avatar
Pirlo's Beard
FISO Jedi Knight
Posts: 20554
Joined: 21 Aug 2013, 17:48

Re: Dod's Blog

Post by Pirlo's Beard »

dod wrote: 16 Jul 2019, 02:07 I haven't watched the film or read the Jon Krakauer book but I might do in future as I really enjoyed 'Under the Banner of Heaven' by the same author.
Definitely watch the film if you get the chance - it's an extraordinary story, beautifully told. Emile Hirsch is terrific and there's a heartbreaking performance from the great Hal Holbrook. It might not be saying much but it's probably the best thing Sean Penn has ever done, the pretentious gobshite.

User avatar
Sutter Kane
Dumbledore
Posts: 7522
Joined: 05 Aug 2010, 12:13
FS Record: Unknown.

Re: Dod's Blog

Post by Sutter Kane »

dod wrote: 16 Jul 2019, 00:08 Whatever else may change between now and the start of the FPL season, one thing I am certain of. My budget will be 99.5m :wink: .
Me too. It's the very definition of flexibility.

User avatar
raoul
Dumbledore
Posts: 5031
Joined: 13 Oct 2005, 18:32
FS Record: Tied for OR 1st worldwide in FPL in GW5 in 21/22 (only to then finish 1m), and ranked 1st in TFF late in season 21/22 (eventual finish 95th). Won an old game called Football Fantastic. Best Fanteam finish 117th and cashed £150. Best FPL finish 2.4K.

Re: Dod's Blog

Post by raoul »

Maybe my reaction is a little too hasty. But.

As you have noted, a 0.1 change in price at the start of the season is a pain because with prices all in 0.5 blocks a 0.1 change is effectively a 0.5 change.

So you hold back 0.5 to allow others to sample the red berries you dared not risk. Turns out they are scrummy and red berries are now 0.1 more expensive than they were. If those berries had been 0.5 higher than the player you picked instead, then they are now 0.6 higher and out of reach. Totally agree that players of the same price who increase are still affordable, but the extra players who are 0.5m higher only stay affordable if they do not go up in price, which may well indicate that you would not want to buy them anyway (a little simplistic that final statement of course).

Does this mean that a 1m safety net is the flex point, because it guarantees the batch of players who are 0.5m above who you chose at any position are all affordable replacements, and some of those who are 1m dearer will also stay affordable for a while?

Of course, whether you can afford to keep 1m back is another story...

User avatar
Patrician
Grumpy Old Man
Posts: 1492
Joined: 13 Oct 2005, 18:32
Location: ...bears no responsibility for bad points forecasts
FS Record: 2nd TFFE 2008, FPL 3 top 10k finishes
Contact:

Re: Dod's Blog

Post by Patrician »

I am probably going to do the same, for a specific reason. I quite want to start with Auba for the first two games, but I am worried that Kane could start well and I might want him for GW3. If he hauls against Villa (very possible) he could rise in price even though he has City in GW2.

That said, lets talk about ...

The accumulation of marginal drains 8-)

I have observed that you get quite wedded to your pet theory. Last season it was the concept of volatility. A fine concept, which helped me win a bet making me £500 better off (huge DGW TC Aguero swing that dug me out of an impossible situation). You however spent a whole season making volatility plays, and in self-reflection you identified this as problematic. This season it might be your quest for flexibility that comes back to haunt you.

Flexibility is not an outcome in itself, and it comes at a cost to xPts. Sure, 0.5M is not much, but it isn't the only flexibility plan you have. You also insist on picking players from the most flexible dead zone price points. You could also have the idea to never hold three players from a team, as it blocks you jumping on another bandwagon in that team in a single transfer, or you could delay/avoid a great transfer because the position is less flexible...and so on.

Each of these flexibility decisions may be a marginal drain on your xPts. Individually they seem small, together they could add up to 3pts per week or more.

Game theory concepts like volatility and flexibility are tools, not outcomes. I prefer to use them for specific reasons rather than apply them as principles.

User avatar
blahblah
FISO Viscount
Posts: 108502
Joined: 13 Oct 2005, 18:46
Location: .. he thinks that he knows something which he doesn't, whereas I am quite concious of my ignorance.

Re: Dod's Blog

Post by blahblah »

That reads more like a marginal vortex....

User avatar
Sutter Kane
Dumbledore
Posts: 7522
Joined: 05 Aug 2010, 12:13
FS Record: Unknown.

Re: Dod's Blog

Post by Sutter Kane »

Yes if you purposely do it every week, it's a drain of course. But to start with 0.5mn does give more flexibility at very little cost at all. It's after GW1 that you may want the maximum amount of flex with only 1 free transfer. The 0.5mn can get you to a price point containing a host of potential early bandwagons.

If you don't leave any money in the bank and go for price points allowing another method of flex, then you may not want to remove a player to get the player you want. This is where flex falls down.

There are indeed pluses and negatives, but at the start, I can't see any negatives. After the first WC, I won't be thinking of flexibility too much as it's over-egged imo; 2 transfers and you can do most things.

User avatar
dod
FISOhead
Posts: 734
Joined: 31 Jul 2015, 23:49
FS Record: I have never won FPL.

Re: Dod's Blog

Post by dod »

raoul wrote: 16 Jul 2019, 08:03 Maybe my reaction is a little too hasty. But.

As you have noted, a 0.1 change in price at the start of the season is a pain because with prices all in 0.5 blocks a 0.1 change is effectively a 0.5 change.

So you hold back 0.5 to allow others to sample the red berries you dared not risk. Turns out they are scrummy and red berries are now 0.1 more expensive than they were. If those berries had been 0.5 higher than the player you picked instead, then they are now 0.6 higher and out of reach. Totally agree that players of the same price who increase are still affordable, but the extra players who are 0.5m higher only stay affordable if they do not go up in price, which may well indicate that you would not want to buy them anyway (a little simplistic that final statement of course).

Does this mean that a 1m safety net is the flex point, because it guarantees the batch of players who are 0.5m above who you chose at any position are all affordable replacements, and some of those who are 1m dearer will also stay affordable for a while?

Of course, whether you can afford to keep 1m back is another story...
I think leaving 1m ITB may be overly cautious. There is a law of diminishing returns in operation here. I am pretty certain I will get better value out of the 0.5m I leave ITB than I would if I put it into my GW1 team. Given that I am already proposing a very conservative opening squad it's a bit of a belt and braces approach anyway. Leaving 1m ITB feels more like belt and braces and gaffer tape.

I'd equate it to choosing the subs bench for your GW1 squad. You definitely want at least one playing sub. Two would be nice but may not be necessary. If you think you need three you probably have volcano insurance and check under your bed for monsters every night before going to sleep.

User avatar
dod
FISOhead
Posts: 734
Joined: 31 Jul 2015, 23:49
FS Record: I have never won FPL.

Re: Dod's Blog

Post by dod »

Patrician wrote: 16 Jul 2019, 08:04 I am probably going to do the same, for a specific reason. I quite want to start with Auba for the first two games, but I am worried that Kane could start well and I might want him for GW3. If he hauls against Villa (very possible) he could rise in price even though he has City in GW2.

That said, lets talk about ...

The accumulation of marginal drains 8-)

I have observed that you get quite wedded to your pet theory. Last season it was the concept of volatility. A fine concept, which helped me win a bet making me £500 better off (huge DGW TC Aguero swing that dug me out of an impossible situation). You however spent a whole season making volatility plays, and in self-reflection you identified this as problematic. This season it might be your quest for flexibility that comes back to haunt you.

Flexibility is not an outcome in itself, and it comes at a cost to xPts. Sure, 0.5M is not much, but it isn't the only flexibility plan you have. You also insist on picking players from the most flexible dead zone price points. You could also have the idea to never hold three players from a team, as it blocks you jumping on another bandwagon in that team in a single transfer, or you could delay/avoid a great transfer because the position is less flexible...and so on.

Each of these flexibility decisions may be a marginal drain on your xPts. Individually they seem small, together they could add up to 3pts per week or more.

Game theory concepts like volatility and flexibility are tools, not outcomes. I prefer to use them for specific reasons rather than apply them as principles.
I like your "accumulation of marginal drains" (I see what you did there :wink:) idea. I think your estimate that they could cost "up to 3pts per week or more" is on the high side though even if (as I assume you are) you are totally ignoring any positives on the other side of the equation. That's a lot! :shock: It's over 110 pts per season. While I agree that there may be costs before benefits of pursuing flexibility I would put that estimate far lower. I'd be interested in seeing how you came to this number.

Another clarification I feel is necessary is that I don't have a problem with owning three players from one team. There is a certain loss of flexibility but you won't be able to get every player you want with a single transfer no matter how much flexibility you build into your team.

For example I have no issue with picking 3 LIV defenders but I would have an issue of picking Origi or Henderson for one of your LIV spots even if they were good value at their price. Similarly if you are choosing a bench player then it makes no sense to choose one from a team where you might possibly want 3 of their players. I would not consider Stekelenburg, Skipp or Nketiah for example. The risk that they would block a transfer I would want to make is low but still far higher than any potential reward from owning them at this stage of the season.

There is another misconception in your post that I wish to address but it's a broader issue that will need a longer theory post of it's own so I'll leave it for now. 8-)

View Latest: 1 Day View Your posts
Post Reply

Return to “FPL Team Diaries & RMTs”