Opinion Questions For Rowsus

Status
Not open for further replies.

Darkie

Leadership Group
Joined
12 Apr 2014
Messages
25,410
Likes
65,496
AFL Club
Collingwood
I think that's a typo on the wiki you were looking at, I think the actor's real name is Pat McCaffrie.
https://www.imdb.com/name/nm0564562/

Try this instead (where I assume the name has come from):
https://en.wikipedia.org/wiki/Hitman_(franchise)

Keep up the good work with this thread, I'm happy to just read along but if I come up with any good questions for you I'll be sure to post them.
I tried to Google the name and this is the closest thing that came up! Figured it must be really obscure ?
 

Connoisseur

Leadership Group
Joined
3 Jul 2017
Messages
38,963
Likes
126,630
AFL Club
Port Adelaide
Not sure of an error Rowsus. Agent 47 name in the show was Pat McCaffrig, played by McCaffrie. Grew up on that show, highly entertaining.
I wasn't aware they had made a typo in the credits. Makes more sense now.

I used to like Get Smart but can't remember which was 47 from that many years ago.
https://www.gilbertpodcast.com/barbara-feldon/
“Agent 99” herself, the funny and utterly charming Barbara Feldon invites Gilbert and Frank to her Manhattan townhouse to share warm memories of “Get Smart” co-stars Don Adams, Ed Platt and Bernie Kopell and gives us her take on the Steve Carell feature film version. Also, Barbara looks back on working with everyone from Dean Martin to Bruce Dern and tells us how she managed to win $64,000 on a quiz show. Plus: Gilbert channels John McGiver! A live rendition of the “99” song! Barbara auditions to be a stripper! And the worst TV movie ever made!

https://www.gilbertpodcast.com/buck-henry/
Gilbert and Frank welcome one of their favorite funnymen, legendary writer-actor-director Buck Henry, who looks back on his 50+ year career and shares hilarious anecdotes about Orson Welles, James Mason, John Belushi and Jonathan Winters (among others). Also: Buck adapts “Catch-22,” praises Richard Benjamin, invents the Cone of Silence, co-directs “Heaven Can Wait” and confirms the Pat McCormick helicopter story. PLUS: “Captain Nice”! “Samurai Delicatessen”! Claude Rains speaks! The hoaxes of Alan Abel! And Buck remembers “That Was the Week That Was”!

https://www.gilbertpodcast.com/bernie-kopell/
One of the funniest and most beloved character actors of his generation, Bernie Kopell joins Gilbert and Frank to reminisce about his six decades in show business, working with legends Steve Allen, Jack Benny and Phil Slivers and his signature roles on Get Smart, When Things Were Rotten and The Love Boat. Also, Charles Boyer apologizes, Raymond Burr takes a seat, Harvey Korman peddles encyclopedias and Bernie remembers his old pal **** Gautier. PLUS: The world’s slowest agent! Louis Armstrong hails a cab! Jonathan Winters lays down the law! In praise of **** Van Dyke (and Mary Tyler Moore)! And a surprise guest calls in to the show!
 
Joined
31 Mar 2014
Messages
2,909
Likes
2,122
AFL Club
Brisbane
Hey d1ck, good to have you back.
Looking at SPS's figures, it looks like he started on the wing, then progressively moved to the back flank. I'm judging this by his I50 numbers starting higher than R50 numbers, but that trend started to reverse about Round 17 on, and also in Round 11.
I'm a bit worried he only averaged 19 Disposals, in a back line that should have seen plenty of the ball. I'm also worried that both his DE and CPR (Contested Possession Rate) are a little down. You'll usually accept a lower than normal CPR, if the DE is a bit higher, but when they're both a down a little on what you hoped, it is concerning. His DE was 70%, and his CPR was 42%. Neither of them terrible, but both sort of in no man's land. Particularly his DE. His TOG was around 79%, which he can probably bump a little, if he plays HBF. He did make some progression last season (20/68, 22/67, 22/82), but he seems awkwardly priced now ($443,300). With Docherty coming back, it's hard to be confident that SPS can bump his numbers enough, to push into that 90+ area that you'd want/need. Certainly being a 4th year player, with 64 games under his belt, I can understand some Coaches hoping he progresses from here. I think he might be more trap, than a happy pick. Sorry.
Grassias my friend! Great to be back and listen to your statistical insight!
 
Joined
26 Jun 2019
Messages
2,559
Likes
9,517
AFL Club
Richmond
Howdy @Rowsus I'm enjoying this thread and really appreciate the time and effort that you put in, the information is invaluable.

I posted this in the Player X v's Player Y thread and thought that your skills could provide an answer for me.

-----------------------------------------------------------------------------------------------------------------------------------------------

By taking a risk with Jackson it would allow me to have:

Roberton ..............D6
Lyons........................M6
Hill............................F4

Leaving only 3 first year rookies on the field (2 if I want to start Cockatoo) and only 6 in total if I class Hill as a rookie.

Banking on Jackson getting 5 games and gaining a bit of value, whilst Gawn might (should) drop a bit over the first 5 if he's not match fit.

With Lyons, I'd be gambling on him gaining some $ considering his finish to last year and then trading him to a rookie on the bubble. Then trade Jackson > Gawn.

For that scenario to work for me I would have to be at least 300k better off. I don't know what each would have to score to generate that outcome.
 
Joined
8 Aug 2012
Messages
418
Likes
2,699
AFL Club
Bulldogs
I will 100% guarantee you, that every single year of SC, the "optimum team" will be full of Midpricers. Mathematically, because of the points/dollar basis, it has to be. Pretty much every year, or close to it, you can construct a set and forget team, that would win SC without even making a trade! That doesn't mean that approach is your best chance of winning SC in any given year.
The best comparison I can give you is, the chances of landing on that "optimum team" are less than the chances of you winning Tattslotto with one entry containing 6 numbers. Midpricers have a success rate of well less than 30%, and there are a hell of a lot more than them, than genuine Prems, who have a success rate (depending on your definition on both Prem and success) of around 60%. The chances of you picking a absolutely brilliant G'n'R team, are about the same as correctly naming two cards drawn randomly from a deck of playing cards. I much prefer that, to the Tattslotto odds.



Thanks for the link. :)
I've seen a similar breakdown of the deck of cards before.
On a similar theme, around 15-20 years ago, I made a game that was a cross between chinese checkers and chess. It was for 2-6 players, and each player started with 6 playing pieces. Those playing pieces had set starting positions on the board, but could be placed in any order by the player using them.
When 6 people played there were 1.43 x 10^45 ways of setting up the board!
When 2 people played, there was 1.13 x 10^15 ways of setting up the board.
For a comparison, the number of ATOMS on Earth is around 1 x 10^50!
 
Joined
8 Aug 2012
Messages
418
Likes
2,699
AFL Club
Bulldogs
Hi again
Following on from your comment that the optimum SC team is likely to be full of mid-pricers.
I think this is probably the best way to frame the discussion.
I get the impression most correspondents just think that mid-pricers are inherently bad and that its suicide to pick them.
But you are saying (and I agree) that getting in a mid-pricer that comes off is inherently a good thing.
In that way, the issue is framed as a "degree of difficulty" decision and I'm quite happy with that!
 
Joined
17 Jan 2015
Messages
1,109
Likes
1,700
AFL Club
Collingwood
Hi again
Following on from your comment that the optimum SC team is likely to be full of mid-pricers.
I think this is probably the best way to frame the discussion.
I get the impression most correspondents just think that mid-pricers are inherently bad and that its suicide to pick them.
But you are saying (and I agree) that getting in a mid-pricer that comes off is inherently a good thing.
In that way, the issue is framed as a "degree of difficulty" decision and I'm quite happy with that!
I think the real message was that it was borderline impossible to predict the combination from millions of options. Having a cookie-cutter team with 1-2 'good' mid-pricers is difficult enough to achieve without trying to get them all.
SC is a stock game, buy low sell high etc. Picking the right captain and trading well though will make more of a difference than nailing the starting team imo
 
Joined
14 Jun 2013
Messages
1,781
Likes
2,685
AFL Club
Melbourne
Hey mate, hope you've been well and enjoyed the off-season to date. My teams been pretty stable outside of my 5 starting gun mids, which I seem to change every day.

I know you're keen on Danger, and he's the one who is locked for me also, but what are your thoughts on the other 600k + mids?

Macrae, Neale and Cripps seem safe picks, Fyfe and Kelly obviously have the scoring potential with an injury risk, Titch is coming back from injury and Bont and Dunkley could either be the best or worst of the lot.

History shows that not all will be good picks but it's tough at the moment to work out who will be and who won't!
 

Rowsus

Statistician
Joined
19 Mar 2012
Messages
29,134
Likes
64,904
AFL Club
Melbourne
HI Rowsus,

I am contemplating Snelling, not sure why. Any observations?
Hi Erek,
to be totally honest, I hadn't given him a second thought, until your question.
Snelling (Fwd $321,800) isn't totally without merit!
His SANFL stats, before getting a mid-season draft spot last year, were really impressive! In the month before he got drafted, he averaged 36 Disposals, 13 Tackles and 8 Clearances. Those are impressive numbers! His VFL stats after getting drafted were pretty good too (27 Disposals, 6 Tackles).
Including that one final, he played 4 games for Essendon last year, for scores of 49, 83, 122 & 59 (in the final). That gave him an in season average of 85. On that number, he is underpriced.
It's a tricky price point. He pretty much has to average close to 90 to be considered a successful pick. In 2019 90 was actually an ok F6 Keeper score. Most years you want a little better, but who knows? 90 might be ok again in 2020!
There are lots of players around that price, that tempt us into often terrible mistakes, but he's one to keep an eye on in the Marsh. If he's getting reasonable mid time, he's probably worth a risky pick. If he's not, he's likely to suffer from small forward syndrome (he's only 176cm).
Well spotted, Erek. I will be watching him with interest!
 

Rowsus

Statistician
Joined
19 Mar 2012
Messages
29,134
Likes
64,904
AFL Club
Melbourne
Howdy @Rowsus I'm enjoying this thread and really appreciate the time and effort that you put in, the information is invaluable.

I posted this in the Player X v's Player Y thread and thought that your skills could provide an answer for me.

-----------------------------------------------------------------------------------------------------------------------------------------------

By taking a risk with Jackson it would allow me to have:

Roberton ..............D6
Lyons........................M6
Hill............................F4

Leaving only 3 first year rookies on the field (2 if I want to start Cockatoo) and only 6 in total if I class Hill as a rookie.

Banking on Jackson getting 5 games and gaining a bit of value, whilst Gawn might (should) drop a bit over the first 5 if he's not match fit.

With Lyons, I'd be gambling on him gaining some $ considering his finish to last year and then trading him to a rookie on the bubble. Then trade Jackson > Gawn.

For that scenario to work for me I would have to be at least 300k better off. I don't know what each would have to score to generate that outcome.
Howdy OnTopBar,
I'm glad you are enjoying my thread.
So to precis your scenario, you want Jackson ($198,300) to play 5 games, Gawn ($697,100) and Lyon ($535,400) to play X games, and end up roughly $300k better off, taking into account Jackson's growth, Lyons potential growth, and Gawn's price drop.
As you are probably aware, I am a Melbourne supporter. Right up front I'm going to tell you, right now for Jackson to even have a good chance to play 5 games early on, it probably means Gawn is missing a number of those games, so it sort counters your plan a little. But let's assume Jackson does play 5, and Gawn also plays.

Jackson (Ruck $198,300) would be doing very well to average 75 over his assumed 5 games. That would have him priced at $285,100. That gives you $86,800 towards your $300k. So you want a combined $213,200 from Lyons up, and Gawn down.
Let's divide that into quarters, and come up with 3 scenarios, and ssume you are willing to wait 9 games for this to happen .
Gawn -$53,300, Lyons +$159,900 - Gawn needs to average 123, Lyon needs to average 145
Gawn -$106,600, Lyons +$106,600 - Gawn needs to average 109, Lyon needs to average 132
Gawn -$159,900 , Lyons +$53,300 - Gawn needs to average 96, Lyon needs to average 119
So, in round numbers, to get your wish, you need Lyons to average 23 more than Gawn, over the 9 Rounds.

Plans like this sound good in your head, but when the time comes, are really hard to execute. If Lyon is averaging 120-125+ after 9 Rounds, are you really going to trade him out? If Gawn is only averaging 100 after 9 Rounds, are you really going to trade him in? It's easy to say yes now. It's really counter intuitive, which is something I generally like and applaud, but it's a tough move to make, when the time comes to pull the trigger.
 

Rowsus

Statistician
Joined
19 Mar 2012
Messages
29,134
Likes
64,904
AFL Club
Melbourne
Hi again
Following on from your comment that the optimum SC team is likely to be full of mid-pricers.
I think this is probably the best way to frame the discussion.
I get the impression most correspondents just think that mid-pricers are inherently bad and that its suicide to pick them.
But you are saying (and I agree) that getting in a mid-pricer that comes off is inherently a good thing.
In that way, the issue is framed as a "degree of difficulty" decision and I'm quite happy with that!
I think the real message was that it was borderline impossible to predict the combination from millions of options. Having a cookie-cutter team with 1-2 'good' mid-pricers is difficult enough to achieve without trying to get them all.
SC is a stock game, buy low sell high etc. Picking the right captain and trading well though will make more of a difference than nailing the starting team imo
Hi again,
There are Midpricers, and then there are Midpricers.
The large percentage of them are very VERY speculative, and then there are the calculated risk types.
Example Aaron Sandilands 2014.
Opened $310,700 and played a 21/108 season. It was the 4th best Ruck average for the season, and 2 of the 3 averages above him only played 17 & 12 games for the season. He had the 2nd best total points scored for a Ruck with 2,269 points, behind Jacobs 2,538, but ahead of Goldy's 2,244 (21/106). Sandi was coming off seasons of 7/64, 12/113, 13/111. His game counts were horrible, and his most recent season also had bad scoring (with excuses). He probably needed to average 88-90, and string together at least 6 or 7 games, to be called a successful pick. Obviously his most recent season caused many people concern, and he started in, from memory, around 45-50% of teams. While he was lower risk, than taking a similar priced 3rd/4th year Def, at a similar price, there was still a pretty high risk factor. The only potential cover was Tom Derickx, who didn't play Round 1, but did play Rounds 2 to 8, and was pretty much the only Rookie Ruck to play that early in 2014. You had to be lucky to pick him. He didn't play Round 1!!!
Let's call Sandi a 40-50% chance of being a good pick. You don't find many midpricers that have that sort of hope of being a good pick, before Round 1, and nearly all of them are injured Prems, with more than 1 year of good scoring, making a comeback. There isn't bundles of players fitting that description in any given season.
Let's say you start a team with 6 Midpricers, which is well short of the optimum team number of Midpricers. Let's say 2 of them fit the Sandi decription, and the other 4 are more speculative. Give the Sandi types a 45% chance of success, and the speculatives a 20% chance of success. Now I suggest to you, that 20% is being ridiculously generous. Check this link to see why:

Questions For Rowsus, page 158, post #3,143

When you are playing Midpricers, you need around 50% of them (or a little better) to succeed, just to break even. If you can get that % up to 60-65%+, you are probably in front. So you have 2 x 45% chances, and 4 x 20% chances. What are the odds you will end up with 3+ successful picks?

Chances of 0 successful picks = 12.39%
Chances of 1 successful pick = 32.67%
Chances of 2 successful picks = 33.21%
Chances of 3 successful picks = 16.67%
Chances of 4+ successful picks = 5.06%

So if you have the 6 Midpricers as described above, you are a 78.27% chance of regretting it, a 16.67% chance of breaking even, and 5.06 chance, of calling it a success. Please remember, a 20% success rate for these speculative picks is incredibly generous. If you drop the speculatives down to 10%, the chance of breaking even is 8.42%, and the chances of success is 1.25%. Success doesn't mean you won SC, it just means those selections were successful, or not, as a group of players in your team.

You may also be interested in:

The Recent History Of Mid Priced Defs

Unfortunately, the tables are blurred, as I used a different method back then, but the summaries are still there.
 
Joined
13 Jan 2015
Messages
833
Likes
647
AFL Club
Brisbane
Whats your view on Peter Ladhams??? I know theres not much if anything to really go off but thought you might think he's good or bad pick anyway

Looks like being a pretty good pick potentially but has only scored well against hopeless or not real ruckmen and reports from their practise match are that he is ahead of Lycett at this stage anyway

Could he be this years Marshall?
 
Joined
8 Aug 2012
Messages
418
Likes
2,699
AFL Club
Bulldogs
Hi again,
There are Midpricers, and then there are Midpricers.
The large percentage of them are very VERY speculative, and then there are the calculated risk types.
Example Aaron Sandilands 2014.
Opened $310,700 and played a 21/108 season. It was the 4th best Ruck average for the season, and 2 of the 3 averages above him only played 17 & 12 games for the season. He had the 2nd best total points scored for a Ruck with 2,269 points, behind Jacobs 2,538, but ahead of Goldy's 2,244 (21/106). Sandi was coming off seasons of 7/64, 12/113, 13/111. His game counts were horrible, and his most recent season also had bad scoring (with excuses). He probably needed to average 88-90, and string together at least 6 or 7 games, to be called a successful pick. Obviously his most recent season caused many people concern, and he started in, from memory, around 45-50% of teams. While he was lower risk, than taking a similar priced 3rd/4th year Def, at a similar price, there was still a pretty high risk factor. The only potential cover was Tom Derickx, who didn't play Round 1, but did play Rounds 2 to 8, and was pretty much the only Rookie Ruck to play that early in 2014. You had to be lucky to pick him. He didn't play Round 1!!!
Let's call Sandi a 40-50% chance of being a good pick. You don't find many midpricers that have that sort of hope of being a good pick, before Round 1, and nearly all of them are injured Prems, with more than 1 year of good scoring, making a comeback. There isn't bundles of players fitting that description in any given season.
Let's say you start a team with 6 Midpricers, which is well short of the optimum team number of Midpricers. Let's say 2 of them fit the Sandi decription, and the other 4 are more speculative. Give the Sandi types a 45% chance of success, and the speculatives a 20% chance of success. Now I suggest to you, that 20% is being ridiculously generous. Check this link to see why:

Questions For Rowsus, page 158, post #3,143

When you are playing Midpricers, you need around 50% of them (or a little better) to succeed, just to break even. If you can get that % up to 60-65%+, you are probably in front. So you have 2 x 45% chances, and 4 x 20% chances. What are the odds you will end up with 3+ successful picks?

Chances of 0 successful picks = 12.39%
Chances of 1 successful pick = 32.67%
Chances of 2 successful picks = 33.21%
Chances of 3 successful picks = 16.67%
Chances of 4+ successful picks = 5.06%

So if you have the 6 Midpricers as described above, you are a 78.27% chance of regretting it, a 16.67% chance of breaking even, and 5.06 chance, of calling it a success. Please remember, a 20% success rate for these speculative picks is incredibly generous. If you drop the speculatives down to 10%, the chance of breaking even is 8.42%, and the chances of success is 1.25%. Success doesn't mean you won SC, it just means those selections were successful, or not, as a group of players in your team.

You may also be interested in:

The Recent History Of Mid Priced Defs

Unfortunately, the tables are blurred, as I used a different method back then, but the summaries are still there.
Thanks Rowsus
Your analysis is interesting and has prompted a few more thoughts and questions.
Firstly if the statistics say there's about a 5% chance of picking 4 or so successful mid-pricers, doesn't that look like a plausible strategy to get ahead of the pack? With 200,000+ coaches in the hunt, the odds of winning are miniscule, a lot of them are setting out on another (G&R) tack, so a POD strategy offering an advantage looks reasonable to me.

Do we know what the start of season price-profile looks like for recent SC winners/place-getters?

You posted some fairly detailed analysis of mid-price defenders and clearly there weren't a heap of big winners on show.
In my thinking there are 2 categories of mid-pricers..
1 - Proven players who have had a bad season due to injuries.
2 - Younger players who are potential improvers

This year there are 44 Defenders priced in the range $400-450K. On my criteria I would look at only 9 of them as follows.
1582069695656.png
What do you think of their prospects and what average would they need to achieve to be rated as a successful pick?
 
Joined
27 Jan 2014
Messages
6,769
Likes
14,766
AFL Club
Fremantle
Thanks Rowsus
Your analysis is interesting and has prompted a few more thoughts and questions.
Firstly if the statistics say there's about a 5% chance of picking 4 or so successful mid-pricers, doesn't that look like a plausible strategy to get ahead of the pack? With 200,000+ coaches in the hunt, the odds of winning are miniscule, a lot of them are setting out on another (G&R) tack, so a POD strategy offering an advantage looks reasonable to me.

Do we know what the start of season price-profile looks like for recent SC winners/place-getters?

You posted some fairly detailed analysis of mid-price defenders and clearly there weren't a heap of big winners on show.
In my thinking there are 2 categories of mid-pricers..
1 - Proven players who have had a bad season due to injuries.
2 - Younger players who are potential improvers

This year there are 44 Defenders priced in the range $400-450K. On my criteria I would look at only 9 of them as follows.
View attachment 15691
What do you think of their prospects and what average would they need to achieve to be rated as a successful pick?
The overall winners often have mid pricers or speculative picks in their starting team. They also often have never had a high finish before. This is the conundrum and your motivations will answer this question better than the data. If your primary motivation is to win the whole thing then the best way to get in front of 200,000 plus other coaches is to gamble on players before everyone else notices them. But also be prepared to crash and burn. Having a sensible team doesn't mean you can't win it and there are very good coaches who follow the game play theory of simply being in contention as often as possible is the best way to play. It just means you are hoping none of the other 200,000 plus coaches have any luck.
Being risk averse is also being luck averse.
 
Joined
26 Jun 2019
Messages
2,559
Likes
9,517
AFL Club
Richmond
Howdy OnTopBar,
I'm glad you are enjoying my thread.
So to precis your scenario, you want Jackson ($198,300) to play 5 games, Gawn ($697,100) and Lyon ($535,400) to play X games, and end up roughly $300k better off, taking into account Jackson's growth, Lyons potential growth, and Gawn's price drop.
As you are probably aware, I am a Melbourne supporter. Right up front I'm going to tell you, right now for Jackson to even have a good chance to play 5 games early on, it probably means Gawn is missing a number of those games, so it sort counters your plan a little. But let's assume Jackson does play 5, and Gawn also plays.

Jackson (Ruck $198,300) would be doing very well to average 75 over his assumed 5 games. That would have him priced at $285,100. That gives you $86,800 towards your $300k. So you want a combined $213,200 from Lyons up, and Gawn down.
Let's divide that into quarters, and come up with 3 scenarios, and ssume you are willing to wait 9 games for this to happen .
Gawn -$53,300, Lyons +$159,900 - Gawn needs to average 123, Lyon needs to average 145
Gawn -$106,600, Lyons +$106,600 - Gawn needs to average 109, Lyon needs to average 132
Gawn -$159,900 , Lyons +$53,300 - Gawn needs to average 96, Lyon needs to average 119
So, in round numbers, to get your wish, you need Lyons to average 23 more than Gawn, over the 9 Rounds.

Plans like this sound good in your head, but when the time comes, are really hard to execute. If Lyon is averaging 120-125+ after 9 Rounds, are you really going to trade him out? If Gawn is only averaging 100 after 9 Rounds, are you really going to trade him in? It's easy to say yes now. It's really counter intuitive, which is something I generally like and applaud, but it's a tough move to make, when the time comes to pull the trigger.
Thank you for taking the time to reply, puts things in a clearer light.

I've been trying to work out the points difference too because that's an important part of the equation, but there are other variables to take into consideration, like cash generation.

Gawn, Steven, Anderson, Young and Starcevich out. ( I think I might have also dropped Bonar > Mead to help facilitate)

Jackson, Ceglar, Dangerfield, Doedee and Roberton in.
 
Status
Not open for further replies.
Top