Mr Positive, sometimes. Posted October 24, 2020 Share Posted October 24, 2020 42 minutes ago, Fife Saint said: 57 minutes ago, Valentino Bolognese said: Normal page: Your login details, as mentioned in the FAQs will be FIRSTNAME-SURNAME-BARCODE. For example, John-Smith-12345678. Your password will be your unique Season Ticket barcode. Cheers, that's the bit I was looking for in my emails but Gmail is pish for searching folders. Think if was posted to us rather than emailed. Might explain why you can't find it. 0 Quote Link to comment Share on other sites More sharing options...
Valentino Bolognese Posted October 24, 2020 Share Posted October 24, 2020 12 hours ago, The Real Saints said: Hello fellow St. Johnstone fans, I've been selected as a wild card contender for a songwriter competition. It's a worldwide contest with loads of entrants and a lovely cash prize amongst many other good things (I've been made redundant recently, so take pity on me). Basically, the top three wildcards will earn a place in the semi-finals. Voting is open NOW and closes on Tuesday at 5pm. If any of you would be so kind as to vote for me ('Ollie Wale - City Lights), I'd be extremely grateful. Thank you. You can vote via the link here: https://www.andrsonuploaded.com/vote-1 Done. Seems a pretty high standard of competition. 1 Quote Link to comment Share on other sites More sharing options...
PauloPerth Posted October 24, 2020 Share Posted October 24, 2020 13 hours ago, RandomGuy. said: Overall McCann May Hendry Wotherspoon Kerr Per90 Hendry Wotherspoon May McCann MOH I know most, if not all, don't care, but pretty chuffed with my rating system, to try and figure out whos contributing and who's not. Thats for last season. Surprised O'Halloran was so high p90, but he outperformed his xG, only Hendry had a better (shot/shots on target) ratio, he won a surprising amount of defensive duels, and of course crushed it as a dribbler. In terms of the P90 ratings, Drey Wright was our least effective player last season... This season I've got... Overall May Wotherspoon Conway MOH McNamara Per90 May MOH Wotherspoon Conway Craig MOH same as above really, stats seem to love him. Kerr the obvious missing player, but hes 6th just behind Craig and not far off the top 5 overall too. Again his missed opportunities to score harm him. Really too early to be fair this season though as players haven't played enough so I'm getting some weird maths. Weighted more towards attackers now, but just slightly. Initially it was the other way, and this time last week I couldn't get away from Ralston/McCart dominating 19/20 and 20/21 rankings. I'm probably not following this very well, but is this not a bit contrived? The original rating system resulted in a list of players high up that you didn't agree with, so you've kept tweaking it until the players you rate are in the top 5. Would it not therefore be easier to just give a list of your top 5 performers by eye rather than by stats? 0 Quote Link to comment Share on other sites More sharing options...
RandomGuy. Posted October 24, 2020 Share Posted October 24, 2020 2 minutes ago, PauloPerth said: I'm probably not following this very well, but is this not a bit contrived? The original rating system resulted in a list of players high up that you didn't agree with, so you've kept tweaking it until the players you rate are in the top 5. Would it not therefore be easier to just give a list of your top 5 performers by eye rather than by stats? I'm hoping to have a rating system that you can transfer to look at replacements, if that makes sense. First way I had it, I had Ralston as our best player last season, which clearly isn't right. It came about because defenders are doing multiple times more "actions" that strikers. So I'm using my eye test to, roughly, predict where I think players should be, and then try and get there with it. I'm not expecting it to be perfect, but I'm at a stage this morning where I've got Ralston lower rated than Tanser, Duffy as our worst CB, and MOH is no longer in the top 5 performers for last season. McCart plays about 60 passes a game, can make about 15 interceptions and win 30 headers. May scores 2 goals, completes 9 passes, and wins 1 header. Its hard to get a balance that, fairly, shows May should be rated higher this season. 0 Quote Link to comment Share on other sites More sharing options...
Valentino Bolognese Posted October 24, 2020 Share Posted October 24, 2020 (edited) 19 minutes ago, RandomGuy. said: I'm hoping to have a rating system that you can transfer to look at replacements, if that makes sense. First way I had it, I had Ralston as our best player last season, which clearly isn't right. It came about because defenders are doing multiple times more "actions" that strikers. So I'm using my eye test to, roughly, predict where I think players should be, and then try and get there with it. I'm not expecting it to be perfect, but I'm at a stage this morning where I've got Ralston lower rated than Tanser, Duffy as our worst CB, and MOH is no longer in the top 5 performers for last season. McCart plays about 60 passes a game, can make about 15 interceptions and win 30 headers. May scores 2 goals, completes 9 passes, and wins 1 header. Its hard to get a balance that, fairly, shows May should be rated higher this season. Would it not then be more sensible not to try to compare the whole team, and have different criteria that relates to the position that they play in? I've not read Moneyball, only seen the film, but I understood that the reason that it was so powerful as a technique was that it completely bypassed the 'eye test' and only used the performance numbers relative to player cost, to understand who delivered the best value (which weren't the players that the scouts went for). Sounds like you're trying something similar but, but does using the eye test to build the model not hard wire in the bias you're trying to avoid? ETA I guess what I'm asking ... are the numbers saying that Ralston was actually a decent enough player, but that was overlooked because he didn't 'look good' to fans? Edited October 24, 2020 by Valentino Bolognese 0 Quote Link to comment Share on other sites More sharing options...
RandomGuy. Posted October 24, 2020 Share Posted October 24, 2020 (edited) 19 minutes ago, RandomGuy. said: I'm hoping to have a rating system that you can transfer to look at replacements, if that makes sense. An example was just looking for potential Wotherspoon replacements further down the line. This guy, with the assist, popped up as the most similar. Free this Summer from Scunthorpe. Played on both flanks and central, and similar per90 figures for everything except aerial duels, and slightly worse defensively. Its nothing more than me just fucking about to see what comes out just now, tbh. Edited October 24, 2020 by RandomGuy. 0 Quote Link to comment Share on other sites More sharing options...
RandomGuy. Posted October 24, 2020 Share Posted October 24, 2020 3 minutes ago, Valentino Bolognese said: Would it not then be more sensible not to try to compare the whole team, and have different criteria that relates to the position that they play in? Sounds like you're trying something similar but, but does using the eye test to build the model not hard wire in the bias you're trying to avoid? I've got individual totals for each section, so can compare players by how good they are defensively etc., I'm just struggling to tie it all together as defenders do far, far, more per game than centre forwards. I think the figures should really just back up what you're seeing, for the most part, so I'm wary of believing a system that told me Ralston was our best player, and Duffy was better than Gordon or McCart. 0 Quote Link to comment Share on other sites More sharing options...
Radford Posted October 24, 2020 Share Posted October 24, 2020 @RandomGuy. pictured after firing up a spreadsheet earlier today. 4 Quote Link to comment Share on other sites More sharing options...
DukDukGoose Posted October 24, 2020 Share Posted October 24, 2020 40 minutes ago, RandomGuy. said: I'm hoping to have a rating system that you can transfer to look at replacements, if that makes sense. First way I had it, I had Ralston as our best player last season, which clearly isn't right. It came about because defenders are doing multiple times more "actions" that strikers. So I'm using my eye test to, roughly, predict where I think players should be, and then try and get there with it. I'm not expecting it to be perfect, but I'm at a stage this morning where I've got Ralston lower rated than Tanser, Duffy as our worst CB, and MOH is no longer in the top 5 performers for last season. McCart plays about 60 passes a game, can make about 15 interceptions and win 30 headers. May scores 2 goals, completes 9 passes, and wins 1 header. Its hard to get a balance that, fairly, shows May should be rated higher this season. It's fair enough having Ralston placed high purely based on stats. That's why teams use stats as a part of the process and not as the main tool. 0 Quote Link to comment Share on other sites More sharing options...
RandomGuy. Posted October 24, 2020 Share Posted October 24, 2020 1 hour ago, Valentino Bolognese said: ETA I guess what I'm asking ... are the numbers saying that Ralston was actually a decent enough player, but that was overlooked because he didn't 'look good' to fans? The same system also has Duffy better than Gordon, which was another massive red flag. I'd say Tanser, Craig and MOH probably are the biggest surprises out, personally, in terms of how high they rank. Its relevant today, but McCann is 2nd bottom in terms of players this season who had played over 800 mins. 0 Quote Link to comment Share on other sites More sharing options...
Mr Positive, sometimes. Posted October 24, 2020 Share Posted October 24, 2020 Shock Spoony is following up a good performance with a distinctly average one. 0 Quote Link to comment Share on other sites More sharing options...
Shaggy Jenkins Posted October 24, 2020 Share Posted October 24, 2020 42 minutes ago, Mr Positive, sometimes. said: Shock Spoony is following up a good performance with a distinctly average one. Honestly if he passes quicker or more decisively on many occasions today he is man if the match. Absolute text book of why people get so fucking angry at him 0 Quote Link to comment Share on other sites More sharing options...
Mr Positive, sometimes. Posted October 24, 2020 Share Posted October 24, 2020 2 minutes ago, Shaggy Jenkins said: Honestly if he passes quicker or more decisively on many occasions today he is man if the match. Absolute text book of why people get so fucking angry at him Just been same the same thing in the match thread and been told I'm talking shite "cause he got in good positions" The difference between winning and not today was the final third decision making. 0 Quote Link to comment Share on other sites More sharing options...
Shaggy Jenkins Posted October 24, 2020 Share Posted October 24, 2020 1 minute ago, Mr Positive, sometimes. said: Just been same the same thing in the match thread and been told I'm talking shite "cause he got in good positions" The difference between winning and not today was the final third decision makin Spot on. Good positions much like good intentions don't pay the mortgage 0 Quote Link to comment Share on other sites More sharing options...
Shaggy Jenkins Posted October 24, 2020 Share Posted October 24, 2020 Though that was Conway s best game for us today however, him and Danny down that side looked very good 0 Quote Link to comment Share on other sites More sharing options...
perthsaint1977 Posted October 24, 2020 Share Posted October 24, 2020 May scores that early chance and I reckon we would have taken them apart. Like plenty other games this season, final ball always the wrong one. Cannot really complain too much as dominated for most parts. 1 Quote Link to comment Share on other sites More sharing options...
RandomGuy. Posted October 24, 2020 Share Posted October 24, 2020 2 minutes ago, Shaggy Jenkins said: Though that was Conway s best game for us today however, him and Danny down that side looked very good Thought the opposite tbh, in terms of the right hand side. The vast majority of the ball was down their side, and the end result was almost always a cross to nobody and/or play slowed entirely down. 0 Quote Link to comment Share on other sites More sharing options...
Mr Positive, sometimes. Posted October 24, 2020 Share Posted October 24, 2020 7 minutes ago, perthsaint1977 said: May scores that early chance and I reckon we would have taken them apart. Like plenty other games this season, final ball always the wrong one. Cannot really complain too much as dominated for most parts. That seems to be the difference in front of goal for us. Very confidence driven. If we get the first goal we'll be hard to catch but the opposite if we lose the first goal though. 1 Quote Link to comment Share on other sites More sharing options...
glenrothes 1884 saint Posted October 24, 2020 Share Posted October 24, 2020 1 hour ago, perthsaint1977 said: May scores that early chance and I reckon we would have taken them apart. Like plenty other games this season, final ball always the wrong one. Cannot really complain too much as dominated for most parts. Should’ve gone across the keeper, and Jason should hit the target too 1 Quote Link to comment Share on other sites More sharing options...
RandomGuy. Posted October 24, 2020 Share Posted October 24, 2020 @Mr Positive, sometimes. tried it the way of doing it per positon, and this how the rankings come out this season... *caveat that players who have played under 700 mins will have unreliable results, as seen for Bryson, simply because every p90 action is so high. Doesn't include todays game. A totally unbiased look at how the players are playing in their respective positions. Kerr wins less headers p90, less defensive duels p90, and looses the ball more in his own half p90, which means he doesn't rank high as a CB. Bryson is off the charts due to a complete lack of minutes, interested to see if he drops right down after today. McCart and Craig have, consistently, come up as our two best players in almost every single go at this that hasn't been personally adjusted. 0 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.