FPL research help please.

FHFC

Thinking about another article for @FFPundits and need a little help.

I know most of you degenerates will have screen grabs or spreadsheet records of previous seasons. If so please post or pm me for 2015/16 season (1) your final OR and (2) how many extra transfers incurring a -4 hit.

You'll find your hits on the season GW history page if you screen grabbed it.

jimmii

I think I have for the last 3 years. I've been doing it for years but never backed them up and lost the previous years when a laptop packed up.

FHFC

jimmii wrote: »

I think I have for the last 3 years. I've been doing it for years but never backed them up and lost the previous years when a laptop packed up.

Just looking for last season actually. Post edited.

jimmii

FHFC wrote: »

Just looking for last season actually. Post edited.

On mobile now but will send it on when I can. When do you need it by ideally?

FHFC

jimmii wrote: »

On mobile now but will send it on when I can. When do you need it by ideally?

No panic. Want to gather as many as I can over next few days or a week and see where it takes me...

Just need the 2 numbers. No need to post screen grabs etc.

jimmii

Have you posted elsewhere? Guessing it will be tough to get a significant sample size from here.

Has your other article gone live yet?

FHFC

jimmii wrote: »

Have you posted elsewhere? Guessing it will be tough to get a significant sample size from here.

Has your other article gone live yet?

Trying twitter fpl heads too.

Other article went up a while back. Link on chips strategy thread.

Benimar

OR 69,359

Points hits - 10 (40 points)

Just as a note, that was my most points hits and worst rank in 5 years in case you are correlating the two!

FHFC

Benimar wrote: »

OR 69,359

Points hits - 10 (40 points)

Just as a note, that was my most points hits and worst rank in 5 years in case you are correlating the two!

Yeah. Thats exactly what I'm gonna do. Which i suspect will show no correlation between less hits and higher rank, but we'll see. I'm working on getting some global data from another source but a small but reasonable random sample would be interesting to graph as a starting point.

jimmii

FHFC wrote: »

Yeah. Thats exactly what I'm gonna do. Which i suspect will show no correlation between less hits and higher rank, but we'll see. I'm working on getting some global data from another source but a small but reasonable random sample would be interesting to graph as a starting point.

Would have thought fpl discovery would have all the info handy enough. They did summary that covered it for 14/15 nothing for last season though.

FHFC

jimmii wrote: »

Would have thought fpl discovery would have all the info handy enough. They did summary that covered it for 14/15 nothing for last season though.

I must have a closer look there too yeah. Thanks.

redzerdrog

Less hits and lower ranking could just correlate to people giving up on the game.

FHFC

redzerdrog wrote: »

Less hits and lower ranking could just correlate to people giving up on the game.

People who gave up on the game don't save screen shots of their season. Or publicise their 2m ranking!

I suspect this will mostly bring out samples from around the top 200 to 300k.

And I suspect there won't be any correlation at all which is kinda the point I'm driving at if proven true.

jimmii

FHFC wrote: »

I must have a closer look there too yeah. Thanks.

I use it most weeks during the season for ownership and captaincy I like how they publish all the data on spreadsheets too.

Ziegler1988

5971
-28 in point hits

TheDazzler

FHFC wrote: »

Yeah. Thats exactly what I'm gonna do. Which i suspect will show no correlation between less hits and higher rank, but we'll see. I'm working on getting some global data from another source but a small but reasonable random sample would be interesting to graph as a starting point.

I think you may see a correlation between less hits and higher rank. And more hits and lower rank.
But if you do, how do you interpret that? I think it's a little bit of a chicken and egg situation.
If people are doing well, they often don't need to take hits (the overall winner last year didn't even use his first wildcard!). The winner the year before that didn't take a single hit all year. If people are doing badly, they may feel they need to take more hits.
Essentially I think 'good' hits, which is extremely hard to quantify, can help your rank and 'bad' hits will hurt it.
In general better players will tend to take less hits than lesser players (as they plan better, have rolled transfers, etc, etc) and when the better players take them, they tend to be better hits than when the lesser players take them (who want the latest 'must have', who haven't saved a rolled transfer, haven't looked at future fixtures, haven't noticed a teams upturn in form, etc, etc.)

I haven't undertaken any research, that's just my feeling on the issue.

iroced

FHFC wrote: »

Thinking about another article for @FFPundits and need a little help.

I know most of you degenerates will have screen grabs or spreadsheet records of previous seasons. If so please post or pm me for 2015/16 season (1) your final OR and (2) how many extra transfers incurring a -4 hit.

You'll find your hits on the season GW history page if you screen grabbed it.

Don't want to be a d*ck but I think it's a meaningless compilation you're attempting.

OR: 69776.
PH: 18.

But my GW33 WC catastrophic disaster triggered 9 pts hits in the remaining 5 GWs. While my 4 first points hits worked a treat in the beginning of the season.


For what you're attempting, you'll need quite some work from each one of us/your contributor, i.e. try and have an objective view of our points hits. First of all point hits for what? Injuries? Suspensions? Bad planning ahead? Taking some punts? Beating price changes? etc... etc... And when you've done it/them, you'd have to look back at who you'd have played/captained/benched had you not gone for it. And it turns to headache if you do successive ones...

I did track mine for a good part of the season last year but I think I did not save it (it's somewhere on this forum).

FHFC

iroced wrote: »

Don't want to be a d*ck but I think it's a meaningless compilation you're attempting.

OR: 69776.
PH: 18.

But my GW33 WC catastrophic disaster triggered 9 pts hits in the remaining 5 GWs. While my 4 first points hits worked a treat in the beginning of the season.


For what you're attempting, you'll need quite some work from each one of us/your contributor, i.e. try and have an objective view of our points hits. First of all point hits for what? Injuries? Suspensions? Bad planning ahead? Taking some punts? Beating price changes? etc... etc... And when you've done it/them, you'd have to look back at who you'd have played/captained/benched had you not gone for it. And it turns to headache if you do successive ones...

I did track mine for a good part of the season last year but I think I did not save it (it's somewhere on this forum).

I think you (and many) are misunderstanding what I'm trying to do. I'm not looking for any particular pattern or to graph any strong conclusion re hits and rank. What I'm really driving at is to disprove the very very common perception that hits are a bad thing, and that never taking them is some sort of badge of honour.

This little exercise i expect to show that some very highly ranked players took lots of hits. And that some very highly ranked players took very few or no hits. And that some poorly ranked players took lots of hits. And that some poorly ranked players too very few or no hits.

So all of ye just gimme your data and shut up!!

iroced

FHFC wrote: »

This little exercise i expect to show that some very highly ranked players took lots of hits. And that some very highly ranked players took very few or no hits. And that some poorly ranked players took lots of hits. And that some poorly ranked players too very few or no hits.

Yeah I got that but we already know that don't we?
Maybe the masses think it's bad to take pts hits and you're gonna shaw some good players do take plenty of hits. Admittedly .

But, I believe the real exercise would/could/should be a deep and thorough analysis of hits. How you can benefit from them when your team is depleted vs spreading your transfers over several GWs. How jumping aboard popular performing players can pay off or not. How deliberately taking hits for DGWs can work a treat or backfire dramatically. How taking multiple short-term punts can destroy your pregression and harm your team. etc... etc...

TheDazzler

FHFC wrote: »

I think you (and many) are misunderstanding what I'm trying to do. I'm not looking for any particular pattern or to graph any strong conclusion re hits and rank. What I'm really driving at is to disprove the very very common perception that hits are a bad thing, and that never taking them is some sort of badge of honour.

This little exercise i expect to show that some very highly ranked players took lots of hits. And that some very highly ranked players took very few or no hits. And that some poorly ranked players took lots of hits. And that some poorly ranked players too very few or no hits.

So all of ye just gimme your data and shut up!!

Ville Ronka took 1 hit and finished 1450th.
I took 7 hits and finished 16551th.

As for what you're trying to prove, I still think it's a very difficult thing to do.
It's very difficult to calculate the gain or loss of any individual hit unless you measure it over 1 GW, which is obviously not the correct way to do it.
Working out that good and bad managers take hits, or don't take hits, doesn't prove anything other than managers take hits. It doesn't prove anything in regard to whether it is mathematically sound to do so. The fact that the debate runs on and on suggests that its cost of -4pts is probably pretty well balanced.

Better managers are better primarily due to having better decision making skills. Weaker managers are weaker primarily due to having weaker decision making skills. And so it follows that better managers are likely to take better hits with a better liklihood of them recouping the -4 and more. And the weaker managers less likely to do so.

The answer is;
It depends.......:)

FHFC

TheDazzler wrote: »

Ville Ronka took 1 hit and finished 1450th.
I took 7 hits and finished 16551th.

As for what you're trying to prove, I still think it's a very difficult thing to do.
It's very difficult to calculate the gain or loss of any individual hit unless you measure it over 1 GW, which is obviously not the correct way to do it.
Working out that good and bad managers take hits, or don't take hits, doesn't prove anything other than managers take hits. It doesn't prove anything in regard to whether it is mathematically sound to do so. The fact that the debate runs on and on suggests that its cost of -4pts is probably pretty well balanced.

Better managers are better primarily due to having better decision making skills. Weaker managers are weaker primarily due to having weaker decision making skills. And so it follows that better managers are likely to take better hits with a better liklihood of them recouping the -4 and more. And the weaker managers less likely to do so.

The answer is;
It depends.......:)

Again. Not what I'm trying to do.

Look at the new season resolutions thread or all around twitter and see how many managers aim is not to take hits.

I'm planning a simple piece about how hits are not the devil and looking for a cross section of data to illustrate it.

I'm not intending to seek a PHD.

El Chapo

OR: 1598
-16 in points hits.

ElTel

OR 8th
12 hits (-48)

TheDazzler

FHFC wrote: »

Again. Not what I'm trying to do.

Look at the new season resolutions thread or all around twitter and see how many managers aim is not to take hits.

I'm planning a simple piece about how hits are not the devil and looking for a cross section of data to illustrate it.

I'm not intending to seek a PHD.

I think I understand. You're trying to prove that hits are not necessarily bad? That's what I was addressing?
I am saying it depends.
It depends on the decision making ability of the manager taking the hit. So it may be good, or it may be bad, depending on their ability, their understanding of game dynamics and the particular situation they are facing.
It's good to show that good managers take hits but if we don't explain to weaker players (who often have a problem with taking too many hits) why, when and to what end the better managers take hits, I'm not sure of the value.

If I may use an analogy;
Some people choose to take drugs.
Some of those people take drugs in moderation, enjoy them and they don't have adverse effects on their lives.
Some people take too many drugs, they become addicted and their lives fall apart.
Telling people that taking drugs isn't inherently bad doesn't help the addict.
If people are going to choose to take drugs, we should advise them on harm minimisation and how to use them responsibly.

So if we're going to tell people it's okay to take hits, we should also tell when, why and to what end we take them.
That's my take on it anyway but we can agree to differ.

Anyway, fair play to you for taking time to undertake some research for the community. I look forward to reading the results.

FHFC

TheDazzler wrote: »

I think I understand. You're trying to prove that hits are not necessarily bad? That's what I was addressing?
I am saying it depends.
It depends on the decision making ability of the manager taking the hit. So it may be good, or it may be bad, depending on their ability, their understanding of game dynamics and the particular situation they are facing.
It's good to show that good managers take hits but if we don't explain to weaker players (who often have a problem with taking too many hits) why, when and to what end the better managers take hits, I'm not sure of the value.

If I may use an analogy;
Some people choose to take drugs.
Some of those people take drugs in moderation, enjoy them and they don't have adverse effects on their lives.
Some people take too many drugs, they become addicted and their lives fall apart.
Telling people that taking drugs isn't inherently bad doesn't help the addict.
If people are going to choose to take drugs, we should advise them on harm minimisation and how to use them responsibly.

So if we're going to tell people it's okay to take hits, we should also tell when, why and to what end we take them.
That's my take on it anyway but we can agree to differ.

Anyway, fair play to you for taking time to undertake some research for the community. I look forward to reading the results.

Didn't intend to get into so much debate here, about an article that I've not even written.

Prove is too strong a word for what I am thinking about. I don't want to empirically or categorically prove anything. I just want to challenge some of the mantras, or stereotypical "rules" that have gone unquestioned to the point that many accept them as fact. I just want to make people think, not tell them another new rule to follow!!

This is to some degree what I was doing with the price changes article, a lot of the "accepted wisdom" about how to gain value from price changes was outdated and hadn't been questioned or objectively analysed.

I believe that trying to have rigid rules, or resolutions about FPL is more often harmful than helpful. And having a rule, that you don't objectively appraise, to generally avoid taking points hits except in extreme circumstances limits you in playing FPL no matter you "skill" or dedication level.

I don't think the drugs analogy is in any way relevant, as taking drugs is not an integral part of life, it's an optional extra, as well as being highly physically addictive. Points hits and using them well and strategically is an integral part of FPL IMO, and opening peoples minds up to that by gently challenging some of the negative rigidity around them is a helpful thing to do I think.

There will be more advice and views on points hits than just the little bit of stats analysis. And as I say I'm not preaching that points hits are great. I just want to suggest that maybe think about it, and objectively assess your situation rather that instinctively avoiding hits "just because".

I have around 20 figures from here and twitter, if I had a few more it would be great, so please if any more of you have the records, please just post you OR and number of hits for last season.

FHFC

To illustrate the above point just today www.fplacademy.com published a really excellent FPL Beginners guide here http://www.fplacademy.com/2016/beginners-guide-to-winning-at-fpl/

And it contains this advice...


ElTel came 8th in the world last year taking 12 hits costing 48 points.

This type of rigid advice to beginners or to anyone is limiting.

Iused2likebusts

I don't keep screenshots sorry fhfc. I agree that there is no right or wrong way to play the game and don't subscribe to the never take hits advice. The problem I see when people talk about hits they have made they don't compare it to what would have happened if they just made 1 transfer. A fair few times you read that someone is up 5 pts after taking an 8 pt hit. E.g hazard 2 to siggy 15 Evans 2 to baines 2 and vardy 2 to Costa 2 . When in reality just doing hazard to siggy would have meant they were up 13 pts.

FHFC

Iused2likebusts wrote: »

I don't keep screenshots sorry fhfc. I agree that there is no right or wrong way to play the game and don't subscribe to the never take hits advice. The problem I see when people talk about hits they have made they don't compare it to what would have happened if they just made 1 transfer. A fair few times you read that someone is up 5 pts after taking an 8 pt hit. E.g hazard 2 to siggy 15 Evans 2 to baines 2 and vardy 2 to Costa 2 . When in reality just doing hazard to siggy would have meant they were up 13 pts.

Thats a good point and one I'll keep in mind in anything I do write.

iroced

FHFC wrote: »

Iused2likebusts wrote: »

The problem I see when people talk about hits they have made they don't compare it to what would have happened if they just made 1 transfer. A fair few times you read that someone is up 5 pts after taking an 8 pt hit. E.g hazard 2 to siggy 15 Evans 2 to baines 2 and vardy 2 to Costa 2 . When in reality just doing hazard to siggy would have meant they were up 13 pts.

Thats a good point and one I'll keep in mind in anything I do write.

Beware though of the hindsight devil. Busts example is an a posteriori view on it. Because we have no info on the context of the transfers. What you really need to know is what were the real options before the GW (e.g. if the poster in Busts example always considered the 8 pts hit like the 3 transfers were bound together, or if he thought of just a -4 or no hit at all). That is THE key. Otherwise you can always read the end product in the best and worst way you want.


I tried to track the efficiency of my transfers/WCs/hits last year in the first half of the season. If you're interested, I'll try to find and link my post to you.

FHFC

iroced wrote: »

Beware though of the hindsight devil. Busts example is an a posteriori view on it. Because we have no info on the context of the transfers. What you really need to know is what were the real options before the GW (e.g. if the poster in Busts example always considered the 8 pts hit like the 3 transfers were bound together, or if he thought of just a -4 or no hit at all). That is THE key. Otherwise you can always read the end product in the best and worst way you want.


I tried to track the efficiency of my transfers/WCs/hits last year in the first half of the season. If you're interested, I'll try to find and link my post to you.

I don't want to get into it that deeply really. Like I say I just want to provoke some thought, and challenge rigidity.

But Busts point of not over valuing a hit with hindsight is worth a mention I think.

Mr.Prodston

Rank: 16,647

11 transfer hits (44 points)

ElTel

The season review thread that Paully D created has this info. (Just in case you forgot.)

http://www.boards.ie/vbulletin/showthread.php?t=2057597845

FHFC

ElTel wrote: »

The season review thread that Paully D created has this info. (Just in case you forgot.)

http://www.boards.ie/vbulletin/showthread.php?t=2057597845

I had. Brilliant.

Have over 40 now and nearly enough to get along with.

Scorpion Sting

It has total transfers but not hits though.

FHFC

Scorpion Sting wrote: »

It has total transfers but not hits though.

Yeah, but I'm happy to assume that the number of times people didn't use a transfer was negligible so the number of hits should be total transfers minus 35.

Scorpion Sting

Okay well then I took 14 hits (-56) for an 8,102 finish.

iroced

FHFC wrote: »

I don't want to get into it that deeply really. Like I say I just want to provoke some thought, and challenge rigidity.

But Busts point of not over valuing a hit with hindsight is worth a mention I think.

Yeah I totally got what you're aiming to do.

But, I was willing to point out that Busts example is incomplete though. Without knowing the player's real options before the GW, we have no idea if his 8 pts hit gained him 5, could have gained him 13 or "lost" him 18 because he highly considered Barkley with no hits who scored 25 pts. I mean, trying to assess whether your transfers/hits worked is only meaningful compared to what were your real options beforehand. Otherwise you can always over or undervalue them.

FHFC

iroced wrote: »

Yeah I totally got what you're aiming to do.

But, I was willing to point out that Busts example is incomplete though. Without knowing the player's real options before the GW, we have no idea if his 8 pts hit gained him 5, could have gained him 13 or "lost" him 18 because he highly considered Barkley with no hits who scored 25 pts. I mean, trying to assess whether your transfers/hits worked is only meaningful compared to what were your real options beforehand. Otherwise you can always over or undervalue them.

Another more global, less time consuming metric which I'm gonna do tomorrow is to plot your GW rank and points hits.

If you're making good decisions on those hits you'd expect to trend upwards in the 3 to 4 weeks after perhaps. Crude but an indicator I think.

FHFC

I finally got round to finishing off this piece of spoofing....seems well received so far.

https://twitter.com/FFPundits/status/767726956284612608

Mr.Prodston

FHFC wrote: »

I finally got round to finishing off this piece of spoofing....seems well received so far.

https://twitter.com/FFPundits/status/767726956284612608

Preaching to the choir for me anyway.

The whole hit philosophy can be boiled down to:

"Oh you don't need to take a hit? That's great, well done on picking the right players at the right time without suffering a Monty Burns Softball losing 8 players in a matter of hours situation.

Although if you do need to rectify a situation then that's okay too. That's why you have the facility to change players with a -4 and not lock in your GW1 team for the season."

I'm also a big fan of the stats deliberately not conforming to a pattern therefore making us right, or not wrong

FHFC

Mr.Prodston wrote: »

Preaching to the choir for me anyway.

The whole hit philosophy can be boiled down to:

"Oh you don't need to take a hit? That's great, well done on picking the right players at the right time without suffering a Monty Burns Softball losing 8 players in a matter of hours situation.

Although if you do need to rectify a situation then that's okay too. That's why you have the facility to change players with a -4 and not lock in your GW1 team for the season."

I'm also a big fan of the stats deliberately not conforming to a pattern therefore making us right, or not wrong

Yeah nothing like using random half assed stats to make a point.

Some guy replied on twitter saying he'd play a wildcard before taking a -4.

Twitter currently full of experts warning people against 'wasting' their wildcard early as if it's a ridiculous thing to do. This forum has more FPL savvy than 99% of those peddling 'expert' advice on twitter (our own @FPLGeneral excepted of course ).