jock _turd Posted October 7 Posted October 7 5 minutes ago, Lone Striker said: Not sure anyone's hating it. Just choosing to ignore it since its meaningless. Just look at Aberdeen. Zero goals scored in 6 matches, but presumably missed some chances. Suddenly score 4 on Saturday and miss some other chances. There's a list of variables which play a part in rendering an xG stat meaningless from one match to another - pitch condition, pitch size, different opposition goalkeeper, different opposition defenders, injuries, more/fewer VAR decisions, weather (high wind, no wind), penalties awarded/not awarded, etc If statistics were really important in football Sunday's game the statistics would tell us we won that game easily... did we win the game easily? Statistics are great if you want to turn something very marginal into something in your favour, I worked in big Pharma and that is how stats work.
Philfigo Posted October 7 Posted October 7 Hahahaha that is brilliant but I'll stick to the eye ball test and that is hooooooooooooof ball. Would be interesting to see what the expected goals for the other teams they have played in each of the games as a comparison to there own expected goals I each of those games. How that can be a positive for them is beyond me, pretty much mean the Bowie Boyle partnership is terrible lol.
Rudy T Posted October 7 Posted October 7 I don’t think it’s a stretch to suggest Hibs have been the worst football team I’ve seen at Tynecastle this season and I include the league cup group. They were ****ing awful. If they’re creating chances it’s because teams are being dragged down to their level and Hibs hoof, fall down, get a free kick routine is getting them into the oppositions box more regularly than other teams. Not a great spectacle on Saturday but we at least gave football a go, an achievement in itself when the balls getting skied 40 yards up the pitch anytime Hibs went near it. I’d be amazed if the Hibs midfielders had more than 20 touches between them. Not only did we win but football actually won in the end. It would be an absolute liberty if they ever claim the flair team mantra again. They wouldn’t look out of place over the road at murrayfield.
WheatfieldWarrior Posted October 7 Posted October 7 11 minutes ago, BigStein said: So basically they miss a lot of sitters? How encouraging... or many of their players are not worth marking....
El Prez Posted October 7 Posted October 7 Xg. Used by desperate IPA drinkers, that help them appear knowledgeable in the pub. Up there with Santa, the Easter Bunny & the Earth being round.
Hømme Posted October 7 Posted October 7 Nothing wrong with xG. I'm sure folk would be jumping in its defence it to back up Hearts. Anyway the more sob stories, false dawns and bigging up of Hibs the better because it keeps Drawvid Gray in a job.
Hømme Posted October 7 Posted October 7 Also, our the clubs full ethos is now about embracing stats and analysing them. It's already paying off but some supporters will still tell us they are a load of pish
Mikey1874 Posted October 7 Posted October 7 A compilation video of all the misses would be helpful here.
NextGenerationJambo Posted October 7 Posted October 7 This thread man. Nearly everything we have access to or take part in in modern day life will have involved/invoked some form of statistics along the way. Take medicine? It'll have went through rigorous clinical trials assessing the effectiveness - using stats. Travelled in a bus/car/train/plane? The materials used and the overall safety profile will have been selected and assessed... using stats. I could go on ad infinitum, every single branch of modern science uses stats to back up hypotheses, because without it relationships cannot be properly quantified as significant. Stats are just a way of mathematically quantifying reality and observations, they are so integral to life, most people won't even be aware of their ubiquity. More relevently, they help sports teams gain an advantage - I mean, the whole premise of JTA is literally built upon this, and Bloom has made billions exploiting it. Those saying it's "pointless", "made up", "nothing compared to the eye test" just have a fundamental misunderstanding of what statistics are, and what they are used for. Which is fine, until you decide to spout a staunch opinion in relation to something you know little about 😂 Statistics is a discipline that's over a thousand years old, and so its fair to say we have refined it quite a bit in that time, but it can still be used improperly. In fact, it often is in order to misrepresent data, and when taken out of context or in isolation they can be misleading. xG by itself, tells us a little, but along with other stats (which are just anaylsis/interpretation of actual events) they help to paint a picture, which is why every team in world football will use them to assess their performance. Really wish people would stop whinging about something that has not only proven itself effective but is also not going anywhere. Not only is it not going anywhere, its use is growing and improving; more and more statistical metrics are being used (see xGOT, xA etc). So you can either embrace it and broaden the scope of your understanding, or continue being perpetually irritated by numbers that simply reflect a degree of reality, whilst outing yourself as someone who doesn't know what they are talking about.
Le Chat Posted October 7 Posted October 7 1 hour ago, Dick Dastardly said: Sitting top of the league in xd (expected dives) as well. Underrated post
Mikey1874 Posted October 7 Posted October 7 5 minutes ago, NextGenerationJambo said: This thread man. Nearly everything we have access to or take part in in modern day life will have involved/invoked some form of statistics along the way. Take medicine? It'll have went through rigorous clinical trials assessing the effectiveness - using stats. Travelled in a bus/car/train/plane? The materials used and the overall safety profile will have been selected and assessed... using stats. I could go on ad infinitum, every single branch of modern science uses stats to back up hypotheses, because without it relationships cannot be properly quantified as significant. Stats are just a way of mathematically quantifying reality and observations, they are so integral to life, most people won't even be aware of their ubiquity. More relevently, they help sports teams gain an advantage - I mean, the whole premise of JTA is literally built upon this, and Bloom has made billions exploiting it. Those saying it's "pointless", "made up", "nothing compared to the eye test" just have a fundamental misunderstanding of what statistics are, and what they are used for. Which is fine, until you decide to spout a staunch opinion in relation to something you know little about 😂 Statistics is a discipline that's over a thousand years old, and so its fair to say we have refined it quite a bit in that time, but it can still be used improperly. In fact, it often is in order to misrepresent data, and when taken out of context or in isolation they can be misleading. xG by itself, tells us a little, but along with other stats (which are just anaylsis/interpretation of actual events) they help to paint a picture, which is why every team in world football will use them to assess their performance. Really wish people would stop whinging about something that has not only proven itself effective but is also not going anywhere. Not only is it not going anywhere, its use is growing and improving; more and more statistical metrics are being used (see xGOT, xA etc). So you can either embrace it and broaden the scope of your understanding, or continue being perpetually irritated by numbers that simply reflect a degree of reality, whilst outing yourself as someone who doesn't know what they are talking about. The key stat is the pace of the ball from Boyle's touch where he hits the ball too hard meaning a chip is his option instead of thrashing it in.
El Prez Posted October 7 Posted October 7 7 minutes ago, NextGenerationJambo said: This thread man. Nearly everything we have access to or take part in in modern day life will have involved/invoked some form of statistics along the way. Take medicine? It'll have went through rigorous clinical trials assessing the effectiveness - using stats. Travelled in a bus/car/train/plane? The materials used and the overall safety profile will have been selected and assessed... using stats. I could go on ad infinitum, every single branch of modern science uses stats to back up hypotheses, because without it relationships cannot be properly quantified as significant. Stats are just a way of mathematically quantifying reality and observations, they are so integral to life, most people won't even be aware of their ubiquity. More relevently, they help sports teams gain an advantage - I mean, the whole premise of JTA is literally built upon this, and Bloom has made billions exploiting it. Those saying it's "pointless", "made up", "nothing compared to the eye test" just have a fundamental misunderstanding of what statistics are, and what they are used for. Which is fine, until you decide to spout a staunch opinion in relation to something you know little about 😂 Statistics is a discipline that's over a thousand years old, and so its fair to say we have refined it quite a bit in that time, but it can still be used improperly. In fact, it often is in order to misrepresent data, and when taken out of context or in isolation they can be misleading. xG by itself, tells us a little, but along with other stats (which are just anaylsis/interpretation of actual events) they help to paint a picture, which is why every team in world football will use them to assess their performance. Really wish people would stop whinging about something that has not only proven itself effective but is also not going anywhere. Not only is it not going anywhere, its use is growing and improving; more and more statistical metrics are being used (see xGOT, xA etc). So you can either embrace it and broaden the scope of your understanding, or continue being perpetually irritated by numbers that simply reflect a degree of reality, whilst outing yourself as someone who doesn't know what they are talking about. You like IPA. That's fine. ❤️
Mikey1874 Posted October 7 Posted October 7 I like the 1 win, 5 draws, 1 defeat stat best. Symmetrical and with a historic reference.
Craig_ Posted October 7 Posted October 7 Pretty sure the photo in the article was from an offside chance too! They really are getting desperate.
I P Knightley Posted October 7 Posted October 7 1 hour ago, Juho_Makela_Goal_Machine said: The outcome of the shot doesn't impact the xG associated with it. And its based on a data set of hundreds of thousands of football games, and for each individual shot, it basically says: for all the shots in our dataset we've seen from this spot on the field, 55% of the time it resulted in a goal, so the xG for this shot is 0.55. That's then added up for each teams shots in a particular game. Do you see a problem here? The big issue I find with xG is that it only rates shots, when it is clear to any football fan who has watched or played the game at any level that not all chances to score end in shots. Take the moment where the ball fell to Shankland in the penalty area on Saturday. If he takes a halfway decent touch, he's got a really good chance of scoring. However it gets away from him and we never get a shot away in the end. The xG for that chance is 0, but if you gave a striker that ball 100 times, they would score maybe 8-10 times, on average? So really we should be saying that that chance gave us 0.08-0.1 xG. The most extreme example is a striker goes through one on one with the keeper and at the last minute takes a poor touch which means the keep can nick the ball of him before he can take a shot. We know that a one on one with the keeper is about a 50/50 chance of scoring, but this would result in an xG of 0. So take xG with a pinch of salt. It can be useful as a barometer, and over enough games it is a reasonably good predictor of a teams underlying performance, but I think we should (and I'm sure Hearts and Jamestown are) look further in depth when analysing single games than just the headline xG score. Thanks I now understand something that I didn't understand a few minutes ago. I've deliberately ignored it up til now as the invention of some plooky nyaffs sitting in darkened rooms with 4 monitors for gaming. I now know that I should ignore it.
Diego10 Posted October 7 Posted October 7 Somebody get on to Jamestown and tell them to stop identifying players please. Nerds
Serge Pizzorno Posted October 7 Posted October 7 Can we get JTA to work out the likelihood of being a Reform voter is based on your opinion on xG? Everything was better back in the day!!!!!!!!!!!!!
NextGenerationJambo Posted October 7 Posted October 7 47 minutes ago, Mikey1874 said: The key stat is the pace of the ball from Boyle's touch where he hits the ball too hard meaning a chip is his option instead of thrashing it in. It's only the key stat if you're assessing that particular chance. The key stat will totally depend on what it is you are looking to anaylse. 48 minutes ago, El Prez said: You like IPA. That's fine. ❤️ Your correlation of IPA drinkers and xG enthusiasts is a form of statistics... 😂
El Prez Posted October 7 Posted October 7 Just now, NextGenerationJambo said: It's only the key stat if you're assessing that particular chance. The key stat will totally depend on what it is you are looking to anaylse. Your correlation of IPA drinkers and xG enthusiasts is a form of statistics... 😂 IPA drinker's shouldn't be included in any metric. Even a pint. My post was only in jest anyway, bud. I actually liked your thought out, slightly condescending post, tbqh.
NextGenerationJambo Posted October 7 Posted October 7 Just now, El Prez said: IPA drinker's shouldn't be included in any metric. Even a pint. My post was only in jest anyway, bud. I actually liked your thought out, slightly condescending post, tbqh. Very good 👏🏼 I know mate, the jest is reciprocal!
El Prez Posted October 7 Posted October 7 Just now, NextGenerationJambo said: Very good 👏🏼 I know mate, the jest is reciprocal! 👍🏻
hereford_hearts Posted October 7 Posted October 7 1 hour ago, HamishMcGonagall said: Absolutely clutching at straws. Reeeeeks of desperately trying to put a positive spin on their season so far, on the back of a derby defeat. "They’re sitting joint second in the rankings by this metric, having carved out 18 really good opportunities in seven Scottish Premiership games. Only Celtic, with 28, have cut opponents open with more frequency than Hibs, with Hearts away down in sixth" Aye, you enjoy your wee make-believe table, while we live over here in the real world with Hearts top of the league. He's very Hibs And how many of Celtic 28 clear cut chances were against the wee team🤔
PhoenixHearts Posted October 7 Posted October 7 (edited) 2 hours ago, Juho_Makela_Goal_Machine said: Not gonna lie, a lot of the xG hate seems to be just old man yells at cloud, back in my day -ism. You don't have to hate something because you don't understand it. Aye alright, specky. Seriously though, I do partially agree. I'm in the u35's age category so not a total dinosaur. Advance stats can sometimes be useful to understand certain trends and tactics in the modern game. I do embrace them to a point, particularly when it suits an argument I'm having. But I tend to ignore them and call them made up new-age woke nonsense if I don't like what they say or if they prove me wrong 🙂 Life is beautiful. Edited October 7 by PhoenixHearts
Juho_Makela_Goal_Machine Posted October 7 Posted October 7 1 hour ago, jock _turd said: If statistics were really important in football Sunday's game the statistics would tell us we won that game easily... did we win the game easily? Statistics are great if you want to turn something very marginal into something in your favour, I worked in big Pharma and that is how stats work. The stats said we had a 39% chance of winning the game, based on the chances both teams created. So no, the stats didn’t say we won the game easily.
Byyy The Light Posted October 7 Posted October 7 Took this from footystats.org. I couldn't read the whole original article because that site is a shitshow but what were they basing their stats on?
Lone Striker Posted October 7 Posted October 7 1 hour ago, NextGenerationJambo said: This thread man. Nearly everything we have access to or take part in in modern day life will have involved/invoked some form of statistics along the way. Take medicine? It'll have went through rigorous clinical trials assessing the effectiveness - using stats. Travelled in a bus/car/train/plane? The materials used and the overall safety profile will have been selected and assessed... using stats. I could go on ad infinitum, every single branch of modern science uses stats to back up hypotheses, because without it relationships cannot be properly quantified as significant. Stats are just a way of mathematically quantifying reality and observations, they are so integral to life, most people won't even be aware of their ubiquity. More relevently, they help sports teams gain an advantage - I mean, the whole premise of JTA is literally built upon this, and Bloom has made billions exploiting it. Those saying it's "pointless", "made up", "nothing compared to the eye test" just have a fundamental misunderstanding of what statistics are, and what they are used for. Which is fine, until you decide to spout a staunch opinion in relation to something you know little about 😂 Statistics is a discipline that's over a thousand years old, and so its fair to say we have refined it quite a bit in that time, but it can still be used improperly. In fact, it often is in order to misrepresent data, and when taken out of context or in isolation they can be misleading. xG by itself, tells us a little, but along with other stats (which are just anaylsis/interpretation of actual events) they help to paint a picture, which is why every team in world football will use them to assess their performance. Really wish people would stop whinging about something that has not only proven itself effective but is also not going anywhere. Not only is it not going anywhere, its use is growing and improving; more and more statistical metrics are being used (see xGOT, xA etc). So you can either embrace it and broaden the scope of your understanding, or continue being perpetually irritated by numbers that simply reflect a degree of reality, whilst outing yourself as someone who doesn't know what they are talking about. Statistics and the methodologies behind what/how & /how often to measure outcomes are indeed extremely useful - and indeed essential - when developing and testing new stuff. in the real world. But the reliability of such stats requires other factors to be controllable throughout all the measurements (e.g. temperature, humidity, distance, etc). No-one on here is arguing against that. Football matches have almost no controllable factors - either within a single match or within a group of matches. Some folk on this thread seem to think that derision of the xG thing must also mean derision of JTA since its based on player stats. Its not - the player stats gathered in JTA cover all aspects of what a player does during each game, and produces searchable summary stats which the user can drill down to see te detail. Pretty sure JTA doesn't contain any notional stuff about whether a player exceeded his xG in a particular game or not, and I'd be worried if Del was even interested in what anyone's xG figure is. xG and JTA stats have no equivalence whatsoever.
Ex member of the SaS Posted October 7 Author Posted October 7 The only metric worth anything is two of our defenders have scored 3 more goals than their striker.
Hackney Hearts Posted October 7 Posted October 7 11 minutes ago, Byyy The Light said: what were they basing their stats on? There's lies, damned lies - and lies.
jock _turd Posted October 7 Posted October 7 (edited) 28 minutes ago, Juho_Makela_Goal_Machine said: The stats said we had a 39% chance of winning the game, based on the chances both teams created. So no, the stats didn’t say we won the game easily. I don't think you have looked at the stats or you are at it. Two times as many shots Three times as many shots on target Three times as many shots inside the opponents box Three times as many successful passes in the final third 61% overall possession... I could go on but reading those stats without seeing the game most football people would say they won that game easily. My point is we did not win it easily and that shows how a set of data can easily be mis read. There is on stat where they excelled backward passes Hearts 69 Hibs HOoooooooooooF only 44 Edited October 7 by jock _turd
Special Agent Dale Cooper Posted October 7 Posted October 7 23 minutes ago, Lone Striker said: Statistics and the methodologies behind what/how & /how often to measure outcomes are indeed extremely useful - and indeed essential - when developing and testing new stuff. in the real world. But the reliability of such stats requires other factors to be controllable throughout all the measurements (e.g. temperature, humidity, distance, etc). No-one on here is arguing against that. Football matches have almost no controllable factors - either within a single match or within a group of matches. Some folk on this thread seem to think that derision of the xG thing must also mean derision of JTA since its based on player stats. Its not - the player stats gathered in JTA cover all aspects of what a player does during each game, and produces searchable summary stats which the user can drill down to see te detail. Pretty sure JTA doesn't contain any notional stuff about whether a player exceeded his xG in a particular game or not, and I'd be worried if Del was even interested in what anyone's xG figure is. xG and JTA stats have no equivalence whatsoever. Top post.
Juho_Makela_Goal_Machine Posted October 7 Posted October 7 9 minutes ago, jock _turd said: I don't think you have looked at the stats or you are at it. Two times as many shots Three times as many shots on target Three times as many shots inside the opponents box Three times as many successful passes in the final third 61% overall possession... I could go on but reading those stats without seeing the game most football people would say they won that game easily. My point is we did not win it easily and that shows how a set of data can easily be mis read. There is on stat where they excelled backward passes Hearts 69 Hibs HOoooooooooooF only 44 All of the stats you’ve mentioned there are far weaker predictors of the result of a match than xG though. And xG says we didn’t win it easily. Not saying xG is a perfect stat by any means but the point is that its not meaningless or made up or woke or whatever (suppose it maybe is woke as most xG shaggers lean that way).
NextGenerationJambo Posted October 7 Posted October 7 7 minutes ago, Lone Striker said: Statistics and the methodologies behind what/how & /how often to measure outcomes are indeed extremely useful - and indeed essential - when developing and testing new stuff. in the real world. But the reliability of such stats requires other factors to be controllable throughout all the measurements (e.g. temperature, humidity, distance, etc). No-one on here is arguing against that. Football matches have almost no controllable factors - either within a single match or within a group of matches. Some folk on this thread seem to think that derision of the xG thing must also mean derision of JTA since its based on player stats. Its not - the player stats gathered in JTA cover all aspects of what a player does during each game, and produces searchable summary stats which the user can drill down to see te detail. Pretty sure JTA doesn't contain any notional stuff about whether a player exceeded his xG in a particular game or not, and I'd be worried if Del was even interested in what anyone's xG figure is. xG and JTA stats have no equivalence whatsoever. Statistics are not just used for new stuff, they are used to assess current stuff, past stuff and even predict future stuff. They are also used to assess many things that have untold numbers of uncontrollable variables/factors: social, political and climate science, to name a few. I am not a statistician, so I am not going to go beyond my scope and pretend to know the advance means by which uncontrollable factors are accounted for, but I do know that they can be, and it is a common occurance in reliable statistical analysis. No one knows what stats JTA are using, so let's not pretend to. All we know is they are far more advanced than the other leading competitors in that field. If one cannot grasp the simple utility of xG, then it is hardly a stretch to assume they are going to understand or appreciate the level of statistical probability that JTA are operating with. As I mentioned previously, xG on it's own only paints a partial picture, anyone who is using it an attempt to paint the full picture, is using it incorrectly and will inevitably run out of paint. I breifly mentioned xGOT (expected goals on target) which is already an enhancement on the xG metric which considers the likihood of a goal based on the initial shot location as well as where abouts on target the shot ends up. No doubt in 5 years time someone will have developed an even more advanced metric that considers the quality of the striker and the goalie (if they haven't already). More and more factors will be considered as the stats become more advanced. People can cry out for the days of old all they want, when football was just football and the only stat that mattered was the end scoreline and the league table, but it will be in vain. Like it or lump it, its the way of the world and the way of modern sports, those who don't embrace it will be left behind. Thankfully our club see the benefit and it's now an integral part of how we operate.
Juho_Makela_Goal_Machine Posted October 7 Posted October 7 29 minutes ago, Lone Striker said: Statistics and the methodologies behind what/how & /how often to measure outcomes are indeed extremely useful - and indeed essential - when developing and testing new stuff. in the real world. But the reliability of such stats requires other factors to be controllable throughout all the measurements (e.g. temperature, humidity, distance, etc). No-one on here is arguing against that. Football matches have almost no controllable factors - either within a single match or within a group of matches. Some folk on this thread seem to think that derision of the xG thing must also mean derision of JTA since its based on player stats. Its not - the player stats gathered in JTA cover all aspects of what a player does during each game, and produces searchable summary stats which the user can drill down to see te detail. Pretty sure JTA doesn't contain any notional stuff about whether a player exceeded his xG in a particular game or not, and I'd be worried if Del was even interested in what anyone's xG figure is. xG and JTA stats have no equivalence whatsoever. So let me get this straight: Stats to analyse and predict player performance - good Stats to analyse and predict team performance - nonsense
Diego10 Posted October 7 Posted October 7 1 minute ago, NextGenerationJambo said: Statistics are not just used for new stuff, they are used to assess current stuff, past stuff and even predict future stuff. They are also used to assess many things that have untold numbers of uncontrollable variables/factors: social, political and climate science, to name a few. I am not a statistician, so I am not going to go beyond my scope and pretend to know the advance means by which uncontrollable factors are accounted for, but I do know that they can be, and it is a common occurance in reliable statistical analysis. No one knows what stats JTA are using, so let's not pretend to. All we know is they are far more advanced than the other leading competitors in that field. If one cannot grasp the simple utility of xG, then it is hardly a stretch to assume they are going to understand or appreciate the level of statistical probability that JTA are operating with. As I mentioned previously, xG on it's own only paints a partial picture, anyone who is using it an attempt to paint the full picture, is using it incorrectly and will inevitably run out of paint. I breifly mentioned xGOT (expected goals on target) which is already an enhancement on the xG metric which considers the likihood of a goal based on the initial shot location as well as where abouts on target the shot ends up. No doubt in 5 years time someone will have developed an even more advanced metric that considers the quality of the striker and the goalie (if they haven't already). More and more factors will be considered as the stats become more advanced. People can cry out for the days of old all they want, when football was just football and the only stat that mattered was the end scoreline and the league table, but it will be in vain. Like it or lump it, its the way of the world and the way of modern sports, those who don't embrace it will be left behind. Thankfully our club see the benefit and it's now an integral part of how we operate. Excellent post
Special Agent Dale Cooper Posted October 7 Posted October 7 3 minutes ago, NextGenerationJambo said: Statistics are not just used for new stuff, they are used to assess current stuff, past stuff and even predict future stuff. They are also used to assess many things that have untold numbers of uncontrollable variables/factors: social, political and climate science, to name a few. I am not a statistician, so I am not going to go beyond my scope and pretend to know the advance means by which uncontrollable factors are accounted for, but I do know that they can be, and it is a common occurance in reliable statistical analysis. No one knows what stats JTA are using, so let's not pretend to. All we know is they are far more advanced than the other leading competitors in that field. If one cannot grasp the simple utility of xG, then it is hardly a stretch to assume they are going to understand or appreciate the level of statistical probability that JTA are operating with. As I mentioned previously, xG on it's own only paints a partial picture, anyone who is using it an attempt to paint the full picture, is using it incorrectly and will inevitably run out of paint. I breifly mentioned xGOT (expected goals on target) which is already an enhancement on the xG metric which considers the likihood of a goal based on the initial shot location as well as where abouts on target the shot ends up. No doubt in 5 years time someone will have developed an even more advanced metric that considers the quality of the striker and the goalie (if they haven't already). More and more factors will be considered as the stats become more advanced. People can cry out for the days of old all they want, when football was just football and the only stat that mattered was the end scoreline and the league table, but it will be in vain. Like it or lump it, its the way of the world and the way of modern sports, those who don't embrace it will be left behind. Thankfully our club see the benefit and it's now an integral part of how we operate. I think most posters in this thread are effectively arguing the same point, that xG in isolation, is not a compelling all encompassing stat. When compiled with other stats, and factors a compelling case can be made but to produce an entire article predicated on a solitary stat is at best lazy and at worst poor journalism.
Diego10 Posted October 7 Posted October 7 2 minutes ago, Special Agent Dale Cooper said: I think most posters in this thread are effectively arguing the same point, that xG in isolation, is not a compelling all encompassing stat. When compiled with other stats, and factors a compelling case can be made but to produce an entire article predicated on a solitary stat is at best lazy and at worst poor journalism. There seems to be a higher percentage who just think it's made up nonsense with no value rather than simply a modelled assessment of the quality of chances made per game. You're bang on with the last point though.
Seymour M Hersh Posted October 7 Posted October 7 I'm told they are top of the xHi statistical league as well. Expected Hibsing it.
151 Posted October 7 Posted October 7 2 hours ago, HamishMcGonagall said: Absolutely clutching at straws. Reeeeeks of desperately trying to put a positive spin on their season so far, on the back of a derby defeat. "They’re sitting joint second in the rankings by this metric, having carved out 18 really good opportunities in seven Scottish Premiership games. Only Celtic, with 28, have cut opponents open with more frequency than Hibs, with Hearts away down in sixth" Aye, you enjoy your wee make-believe table, while we live over here in the real world with Hearts top of the league. He's very Hibs That is some of the most pointless shit ever to get involved with in the game The point of football is to score goals, not create chances where 75 times out of 100 it goes in. Especially not if your strikers can't put them away. Players like Andy Carrol used to always come with the tag "even though he's not scoring - he's getting in to great areas" and that looks like it has evolved into "xg".
Lone Striker Posted October 7 Posted October 7 13 minutes ago, Juho_Makela_Goal_Machine said: So let me get this straight: Stats to analyse and predict player performance - good Stats to analyse and predict team performance - nonsense No, you've not got this straight at all.
Juho_Makela_Goal_Machine Posted October 7 Posted October 7 15 minutes ago, NextGenerationJambo said: Statistics are not just used for new stuff, they are used to assess current stuff, past stuff and even predict future stuff. They are also used to assess many things that have untold numbers of uncontrollable variables/factors: social, political and climate science, to name a few. I am not a statistician, so I am not going to go beyond my scope and pretend to know the advance means by which uncontrollable factors are accounted for, but I do know that they can be, and it is a common occurance in reliable statistical analysis. No one knows what stats JTA are using, so let's not pretend to. All we know is they are far more advanced than the other leading competitors in that field. If one cannot grasp the simple utility of xG, then it is hardly a stretch to assume they are going to understand or appreciate the level of statistical probability that JTA are operating with. As I mentioned previously, xG on it's own only paints a partial picture, anyone who is using it an attempt to paint the full picture, is using it incorrectly and will inevitably run out of paint. I breifly mentioned xGOT (expected goals on target) which is already an enhancement on the xG metric which considers the likihood of a goal based on the initial shot location as well as where abouts on target the shot ends up. No doubt in 5 years time someone will have developed an even more advanced metric that considers the quality of the striker and the goalie (if they haven't already). More and more factors will be considered as the stats become more advanced. People can cry out for the days of old all they want, when football was just football and the only stat that mattered was the end scoreline and the league table, but it will be in vain. Like it or lump it, its the way of the world and the way of modern sports, those who don't embrace it will be left behind. Thankfully our club see the benefit and it's now an integral part of how we operate. You get it lad, so prepared to be called a geek 👍 Geek
jock _turd Posted October 7 Posted October 7 12 minutes ago, Juho_Makela_Goal_Machine said: All of the stats you’ve mentioned there are far weaker predictors of the result of a match than xG though. And xG says we didn’t win it easily. Not saying xG is a perfect stat by any means but the point is that its not meaningless or made up or woke or whatever (suppose it maybe is woke as most xG shaggers lean that way). Did you read what I posted? This is stats from the game if you gave someone xG Hearts vs xG Hibs and asked for a prediction the winner of the game it would be nowhere near as accurate as the set of stats from the game. xG to predict the outcome of a game means nothing without other data, that is a fact.
Juho_Makela_Goal_Machine Posted October 7 Posted October 7 1 minute ago, Lone Striker said: No, you've not got this straight at all. That's what your post said mate. JTA uses player stats to predict and analyse player performances, and that is totally different to using xG to predict and analyse team performance (no equivalence is what you said).
Juho_Makela_Goal_Machine Posted October 7 Posted October 7 4 minutes ago, jock _turd said: Did you read what I posted? This is stats from the game if you gave someone xG Hearts vs xG Hibs and asked for a prediction the winner of the game it would be nowhere near as accurate as the set of stats from the game. xG to predict the outcome of a game means nothing without other data, that is a fact. You're wrong. If you looked at the stats you provided, you're right in saying that you would have thought that Hearts would have beat Hibs comfortably, and I am saying that the xG painted a more accurate picture that it was really close. Most of the 'stats' you're banging on about are shots, and xG is just a more descriptive extrapolation of shots.
NextGenerationJambo Posted October 7 Posted October 7 2 minutes ago, Special Agent Dale Cooper said: I think most posters in this thread are effectively arguing the same point, that xG in isolation, is not a compelling all encompassing stat. When compiled with other stats, and factors a compelling case can be made but to produce an entire article predicated on a solitary stat is at best lazy and at worst poor journalism. Exactly, statistics and statistical anaylsis are two very different things, both as important as each other. Stats are useless without the proper interpretation. So obviously, if you fail to grasp the concept you are either going to think it's pointless, or the be all and end all. Reality, as it usually does, lies somewhere in the middle. I'm happy for journos to continue misinterpreting stats to make hibs out to be a better team than they are, whilst underestimating us.
Hackney Hearts Posted October 7 Posted October 7 4 minutes ago, Juho_Makela_Goal_Machine said: xG is just a more descriptive extrapolation of shots. That is true. "Looking like scoring" often relates to more than just 'shots' though. So xG is more useful than just the raw shots (attempts) stats, but it falls a long way short of painting the whole picture of which team is more likely to score in any given match.
NextGenerationJambo Posted October 7 Posted October 7 20 minutes ago, Juho_Makela_Goal_Machine said: You get it lad, so prepared to be called a geek 👍 Geek I'll accept the label wholeheartedly 😂 I'd rather be a geek about something than ignorant.
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now