About Shom

Shom Biswas is a writer from India. @Spinstripe

FPL Podcast Observation 11: xG? xA? How to get FPL-Stats-knowledgeable

How does one get football-stats-knowledgeable?

A friend at FISO had asked me on how to get into football stats for FPL purposes. This observation is an expansion to my response to him, and is about
1) how one can understand football stats better, and
2) how one can continue to upgrade one’s knowledge every week in the face of this barrage of information.

Basically on how to drink from the firehose of FPL information, and on how to continue to use the firehose well.

My FPL playing style has not been too stats-based historically. I think that I have a decent grip of the concepts of the predominant football stats and their application to FPL. However, I am a bit of a lazy FPL stats person. And while I do not trust either of my gut feel or my eye-test, I have always been a football strategy nerd. I’d go Zonal Marking (@Zonal_Marking) over Double Pivot (@DoublePivotPod), almost everytime.

However, there are two problems here:

  1. I personally do not enjoy poring over FPL stats and numbers over the weekend. That is a substantial part of what I do at work, and I think the moment FPL will become work rather than play, I would either quit FPL or quit my job. The former is much more likely, haha.
  2. There is a particularly demanding two-year old who needs my attention over the weekend. There is also a particularly fierce better half, who does not share my love for football – so time on the telly is rationed. Two matches, that’s all – the Arsenal match and any other. Eye test, therefore, is not always comprehensive.

So I do not have a choice but to delve into stats to keep close to the game.

I have created a shortcut for myself. I listen to about 4-5 hours of podcast every week, while either driving back home from office, or while on the late-evening run. Do you have time for listen to podcasts? Here would be my recommendation on how you can be on top of football stats for FPL.

First, understanding the stats.
I recommend the most recent FML FPL podcast / fireside chat with Michael Caley. It is amazing. Alon is a very smart and knowledgeable FPL manager and host, and Michael Caley is one of the most well-known football stats-man. Below is the link to the pod. It will be an hour well-spent.
Link Here

I would also recommend a detailed post by our resident stats-expert at FISO, and perhaps one of the most consistent FPL managers in the world, Stemania on the xA stats. It gives a very good ‘current-state’ perspective for xA.

Second. Keeping abreast.
If you have the patience to go over tables and stats, become a Fantasy Football Scout member – very comprehensive, very good. I had been a member at FFS previously, but I feel that I have not able to use the stats as well as I would have liked to. This is what I do, and I would recommend. Listen to 2-3 of the following podcasts every week. These should give enough perspective on the stats behind the FPL players and teams. I do not make notes too often, I do not pore over tables. But just by osmosis, I capture enough important stats to make do.

Specific Stats-pods of football leagues and players across the world – with an in-depth look at stats. Nothing to do with FPL. Everything to do with stats.

  1. Double Pivot ( @DoublePivotPod ‏ )
    http://podbay.fm/show/1121866859
  2. StatsBomb ( @StatsBomb )
    https://statsbomb.com/category/podcast/

FPL Oriented ones which primarily are about playing the game, but give regular (near-weekly) stats-based read-outs of Premier League teams and key players

  1. Who got the Assist ( @WGTA_FPL ‏ )
    http://whogottheassist.com/category/podcasts/.
  2. FML FPL ( @FMLFPL )‏
    http://fmlfpl.com/pod
  3. 2 Guys 1 Cup ( @FPL2Guys1Cup )
    https://anchor.fm/FPL2Guys1Cup
  4. It would be remiss of me not to mention Fantasy Football Scout – they have been pioneers and flag-bearers of FPL in many ways. Loads of respect. But I think they are missing a trick by not having a specific-to-stats readout every week. Joe’s (@FFScout_Joe) Goals-imminent table is really crying out for a podcast readout. What a brilliant innovation.
  5. To me, the one and only place of all stats-related discussions is the absolutely peerless yearly STC thread at FISO. Regularly visited by some of the smartest FPL minds I know – many of whom have brilliant FPL records as well. This year is a bit of a down-year: Mav3rick is on a sabbatical (you are missed, Mav), and Stemania has just recently embraced fatherhood (Congratulations!). But it is still excellent.

That has been good enough for me. I listen to podcasts, and I chat on FISO (not so much at twitter yet, but heck, I have just joined in). Perhaps they would be good enough for you too?

Advertisements

FPL 17-18: The Continuing Value of the Expensive Defence

You know that FISO (link here) is my regular haunt, right? Excellent forum, rather old-fashioned, good place for a quiet chat with FPL friends — so, over there, Aldershot Rejects, one of the FPL managers whose style I admire (and top 700 ranker last season), was talking about the trend of the expensive defence which was in vogue at the first few Gameweeks, but has petered off now. I added a very pretentious comment as an addendum, which I would like to share. While pretentious, I think it is also right. But you tell me what you think about it. 

————————–

The power of the big defence is the regular trickle of 6 points that come in from the CS. CS is the bread and butter of the defence.

The big defence, this season, is fulfilling its value. Comfortably – and I have numbers to show. Early season, when the big defence became fashionable suddenly – Most who jumped in did so because of the goal / assist points that Mendy and Alonso were getting. Once the G+A became more scarce, the masses moved out. The regular bread on the table i.e. CSes are still coming in – the kebabs of the G+A are scarce, and the crowds have moved on in search of that juicy steak. As they always do.

Continue to keep faith on the everyday bread, my friend. That reservoir seems to be pretty full still.

Hunter-gatherer is fine no problem, but farmer is great too.

 

Mumbaikar

Yesterday, a star was born.

Not that we did not know about him earlier – he was the star of India’s junior cricket circuit. And the words we would hear from the Bombay media were of genuine appreciation – This kid is the next greatest to have come out of Bombay after Tendulkar. And after Gavaskar, only Tendulkar could be considered better.

Really? Only Gavaskar and Tendulkar?
You are saying … one minute. Hold on. Bombay. Seriously?

Ajinkya Rahane. Wasim Jaffer. Vinod Kambli. Ravi Shastri. The luckless Amol Muzumdar. Sanjay Manjrekar. Rohit Sharma. Heck, the Colonel himself – Dilip Vengsarkar.
You are saying that this kid is potentially better than them all?

That’s what they said. We, from the rest of India hid a smirk – those Mumbaikars always inflate the quality of their own. Sure, he must be good. But he cannot be that good.

He scored a 154-ball 134 in his first test innings. At 18 years of age. While opening the batting. With his partner having departed at 0. And having played technically correct cricketing shots all throughout.

His name is Prithvi Shaw. And you will hear more of him for the next decade and more.

FPL Podcast Observation 10: Always Cheating Ep 143 She’s So Heavy – Be A Rebel! One – more – time !

Let’s hear it for the guys at Always Cheating ( @hailcheaters ). Here is the pod.

Here is the Pod

And just have a read of this exchange.

With increase in available information, is the template getting harder and harder to beat?

Harder to keep up with, I think. The template changes every week!

We are not playing against robots… hold on: What is a template team? A template team is a team made up of players who are doing extremely well (blogger’s edit: and seem to continue to do well in the foreseeable future) – so if the goal is to get players who are doing and will do better than other people – then that is, kind of, THE GAME. Anyway, a true differential only lasts for a week or two – before everybody has them. AWB was a differential for a week! Before everybody figured out that oh, here is a CRD for 4m who is also picking up assists! Same with Matt Doherty – I mean, Matt Doherty will have perhaps 30% ownership within the next month.

 

Excellent. Thanks for saying that. THAT IS THE GAME!

Here’s my point. “Be a rebel” is nice as a bumper sticker, and as a T-shirt slogan for teenagers. However, the inherent purpose for FPL is not to be a rebel – it is to get points. Effectively there are 2 ways of playing the game: 1) make the most sensible decision over and over again and get the maximum benefits out of it. (or) 2) Not really care about the results.

To be a good differential pick – first and foremost – it has to be a good pick. Similarly, the safe pick (say for example, the pick everyone else has made) is not immediately (or necessarily) a good pick.

As you must have realized, I have had a bit of an epiphany after the series of posts from Stemania and Stu255. Fantastic Exchange: Link Here ]. A good pick is a good pick – simple as that. My thoughts on this are still developing, but I seem to agree in principle.

Be-a-Rebel-ing could be a lot fun for a few – and I have no problem with that AT ALL. However, that is not the game.

FPL Podcast Observation 9: Who Got The Assist BE and FPL (w/ Simon March) – Show Us Your Rings

Link: Who Got the Assist Pod 48: BE and FPL

The anecdote from today comes from 2010, when I had just landed at Texas. I was young – six odd years into my professional career. The company I had recently joined, had concluded a PE-backed (and very bitter) hostile takeover of an organization there. I was to project manage a key business transformation activity on a certain division of the acquired organization. The Managing Director that had hired me was an inspirational figure – very abrasive, very impatient for results, but also an innovator, a good listener and quite a humane guy. In our initial strategy workshops – and we had a very heterogeneous team within the transformation group – he had this point of starting with a “Show Us Your Rings” conversation, where we basically start with telling the rest of the group what are the credentials we have to be on the table. It was not a session to judge others by, the MD ensured that – the assumption was that we were hired well, and that different folks would bring in different skill-sets. But this medals-on-the-table introduction would ensure that everyone else would know the point of reference anyone was coming from.

This process worked as a charm. The loudmouths, with their half-baked ideas and boisterous verbosity, would invariably come to terms with the genuine expertise and/or considered opinions of others in certain areas – and would require to sit back and think. Often, the results were startling – as I mentioned, the folks were hired well. As best-practice exchange was key, this led to some excellent collaborative work.

This is important to know before one listens to this podcast, about Behavioral Economics. I have read a book on Cognitive Biases, had taken a sensational course called ‘The Science of Everyday Thinking’ (on EdX- totally recommended), and indeed in private conversation would not hesitate to throw in an opinion or two. But when actual practitioners and researchers at the field are discussing on interpreting their learnings on Fantasy Football (one of whom BLOODY WON FPL a couple years ago) – it would be prudent for me to recommend the entire edition, rather that say such-and-such is my opinion on this issue.

Suggestion – listen to this, even if you do not agree with everything. I kinda sense that I don’t agree with a few elements either, but nonetheless came out a fair bit more knowledgeable.

Especially listen to Simon March’s suggestion from 23rd to the 30th minutes.

 

This is really, really good stuff.

FPL Podcast Observation 8: Fantasy Football Scout Meet the Manager – Phil Ampleford (Philman) – How to Play FPL (Volume 2)?

How to Play FPL – Vol 2

Link to Fantasy Football Scout Meet the Manager – Phil Ampleford (Philman)

This is a brilliant series, guys. I would recommend not to miss a single Meet the Manager. As the most popular FPL advisory forum there is, there is an important purpose that FFS serves. And they do not take the mantle lightly, they do some interesting stuff with it. Like this feature.

So, Phil Ampleford (alias Philsays), the FPL manager in focus for the week, and I have records that are within the same ballpark. His is better and more consistent of course – he is one of the top 100 FPL managers around, and I am not quite there, but in the general vicinity I’d like to assume. I found this out once the ranks were read out.

It is especially interesting to follow a couple of FPL managers every season – especially ones who are making the move to break the glass ceiling and become truly elite. (And that’s not necessarily per FFS HOF ranks – I see the FFS HOF more as an FPL ELO ranking than a true HOF – focused as they are to the more recent performances).

Last season I tracked a few players, and perhaps the most interesting two among them were Anders Lium (#28 in HOF, rank 52 last season) and Simon Rutherford (#33 on HOF, rank 58 last season). Ideally, we can say that both of them were very successful, yes? And these are exactly the kind of players – who with last year’s performance have become truly elite.

Simon’s ( @Heroes_FPL ) was the most radically different in format last season (Vardy + Firmino as strikers. No Kane No Aguero) – it was the most exciting to follow. Not reckless, but unconventional in a very controlled way. Anders’ was the opposite. He caught onto template early, and never veered from template until alternate templates appeared. So much so that I saved him on my bookmarks as “Anders Template – risk-averse”.

I am tracking both of them this season as well – they are playing a similar strategy this season as last, and both are doing very well. Worth a follow. You can pick up the links from the FFS Hall of Fame.

So, back to this episode. Philsays plays a game which is dramatically different from mine. 3-4-3, cheap in defence, early hits and never hit-shy, very price-conscious. Almost the mirror opposite of my normally 4-in-defence, at least two high-priced defenders, very hit-averse, and not very price-conscious style. And perhaps the most illuminating was the discussion on last season, where he finished at the late 90Ks. The reason he gave was that he veered a lot from the style of play that he was comfortable with – and tried to play like a different player. And his rank suffered as a result. He is back to playing in his normal way again, and again doing very well.

So I do not believe that there is a single way (be that risky or safe) to be really, really successful at FPL. Simon, Anders and Phil are all excellent players, and their styles are each quite different from others.

But it is very important to identify the way of playing that one is comfortable with (and that comes with a bit of trial and error), and even more important to not dramatically differ from it ***, once identified. Maybe there will be minor ups and downs for different seasons (luck does play a factor) – but the general average result should be good.

 

 

 

 

*** Of course, this only matters if you consider a decent season-end rank as an important target. Sometimes, just trying out an alternate approach – season-end rank be damned – is a fine reason to play differently as well. That’s okay, you will be back next season. And sometimes even just letting go is fine. It’s okay – FPL has to be fun. Otherwise, what’s the point anyway?

FPL Podcast Observation 7: How to Be Better @ FPL – Fantasy Football Scout #268 and Meet the Manager – Nick (Triggerlips) –

Meet The Manager – Nick (Triggerlips)
https://www.youtube.com/watch?v=hSYscXNHe9s

FFScoutcast #268
https://www.youtube.com/watch?v=Ek9AGlB4o3k

 

These are perhaps the two most intriguing podcasts I heard recently.

First, Nick (Triggerlips), whose twitter handle @NickTriggerlips I follow, was the featured manager for the week. When asked about what has been the biggest change in the game in the last few years – he talked about the tremendous overload of information. And it is true – Everyone seems to know everything. All data is at everyone’s fingertips, new data has been created to represent and explain the game better, all information about all clubs and all players are available now – there is no Carlos Kickaball anymore, and even all analysis has been done on all those data points – all juice has been extracted out of the data. And one does not have to go look for this data – it is available at my laptop screen, and yours. Earlier, there were just these few forums like FFS and FISO, where really only hardcores visited (PS: is it really FY-So? I have always called it Fee So).

Also, listen to the podcast, from the 47th minute onwards. David has made some excellent analysis of how Arsenal’s right wing is a weak spot because Bellerin is always so much high up the pitch (and he does not mention that Mustafi, the covering Right-Central-Defender is prone to a brainf*rt – while the left-sided central defender, Sokratis is quite good positionally, as is Monreal, the left full back) – so it seems ripe for the taking for Richarlison. Similarly Joe speaks about the headed-goal scope for Cardiff.

Do you see how granular data has become? The data has really been beaten to death. We all know. Everything.

There will never be another underpriced defender, or fantastic differential that will remain so after 3-4 gameweeks. We serious FPL players used to think that we would invariably climb the ranks at around the 25th GW or so because the casuals would be drifted out. But that is not the case anymore – or at least, there have been many more serious players than there were previously. Triggerlips mentions that there was a time when getting into the top 10K used to be the Gold-Standard event. And I remember at FISO, there was a thing about being in the Millionaires’ club 9i.e. having a GW rank of more than a million – those were the days when there used to be 1.2-1.5M players. These days, 1.2M is a decent GW rank. And a very decent top 1% is a lowly 60K or so…

So with the availability of all this data, and with the influx of so many good managers – what differentiates a good manager from a bad manager? Nick gives a few tips about it in the Meet the Manager. It is about playing skills. Patience. Not taking unnecessary hits. Being very careful about moving a good player out for another good player. Playing the prices etc.

I think it is equally important to understand about one’s own playing style, and take the effort to absolutely master it.

Are you an aggressive hit-taker? Sure, you might want to be – but are you often being successful taking the hits? Make an analysis, follow the style, and see if you are actually good at it. It’s not good enough to just want to be an aggressive player – one has to be good at it too.

Are you a differential-seeker? Do you have a knack of picking up some lowly-owned players, and consistently taste success with it? Track it for a season or two, and make your impartial judgment around it. If this is not your style, do not embrace this style, and keep to the template picks often.

Are you just risk averse? Maybe you are in real life, but is that something you are good with at FPL also? These are answers you need to find out about yourself. At FISO, we have been tracking the exploits of the legendary, shadowy figure of Ville Ronka for a while now. I am a fan at the altar of Ville, but what impresses me so much about Ville is how extremely comfortable he is about his playing style. Go hell or high water, he would not change his style – and every season, he is up there in the top 1K. In fact, the only season that he came down from his lofty peaks was also instructional. In Season 16-17, at the last few GWs of the season, Ville took hits, made monster changes and chased differentials from a 12k-odd position, and only made the scenario worse for himself, eventually ending with a 21k rank. This was not his game – and even Ville Ronka could not play like (say) Martin Kouwenberg and be successful – while Martin K does it every year and does brilliantly very often.

And not just these playing patterns, it’s also about formations: Are you good at rotating cheap defenders? Are you most comfortable with the 3-4-3? There are so many of these questions that need to be answered for yourself as a Fantasy Football player, before you can understand yourself.

But once you do that, here’s the benefit you can glean from it. There are many ways to skin the cat – there is no one way to do well at FPL. As long as one is comfortable doing one’s own thing – one has a good chance of doing very well indeed. And even if the season is different from others, one would have more flexibility to change format without totally flying off the rails.

This was also something that I have been debating with Ruth_NZ, a FISO colleague, last season. And I think there is good reason to pursue this further in future seasons. Jigyasa is a wonderful thing.