Saturday, March 03, 2007

At What Price Ryan Smyth?

Kinger's got an analysis of Ryan Smyth's worth over on Always On The Road and has set up a collection of strawmen to abuse. He then proceeded to post it at HF and welcome all challengers.

First off, let's give some background on Kinger's approach. It's basically a bunch of comparables and he focuses on counting stats per game. The list is as follows (with cap hit and contract length in parentheses):

Arnott ($4.5MM/yr, 4yrs)
Sullivan ($3.2MM/yr, 2yrs)
Brind'Amour ($3.6MM/yr, 4yrs)
Koivu ($4.75MM/yr, 2yrs)
Alfredsson ($4.34MM/yr, 3yrs plus two mutual options)
St. Louis ($5.25MM/yr, 4yrs)

Alright, first off let's get rid of the useless comparables, shall we?

The Brind'Amour comparable is particularly laughable. Brind'Amour signed his deal at the ripe old age of 35 and is far too old to be an meaningful comparable

Alfredsson's deal is from before the lockout and thus has the benefit of being both stale and reduced by 24%. On top of that, he's got a couple of mutual player/team options at the end of the deal at $3.8MM/yr which drives down the overall value of the contract and thus the cap hit as well.

Although Smyth does compare reasonably well with both those players, their situations aren't really close at all. Forget about those two strawmen.

Since I'm sporting, I'll add a couple of comparables to replace those two. Jokinen and Doan come to mind. If you're keeping score, the new list is:

Arnott ($4.5MM/yr, 4yrs)
Sullivan ($3.2MM/yr, 2yrs)
Jokinen ($5.25MM/yr, 3yrs)
Koivu ($4.75MM/yr, 2yrs)
Doan ($4.50MM/yr, 5yrs)
St. Louis ($5.25MM/yr, 4yrs)

Personally, I'd say the important items for comparison are ES and PP performance, strength of opposition, and distribution of minutes by situation. I'd love to reliably include goal differential at 5V5 and the PPGF/hr numbers (especially the latter, because Smyth has killed there for two years running) but I really just have the numbers for Smyth and the other Oilers for the past couple of seasons. I don't have the Corsi metric (The for/against differential of shots + missed shots + blocked shots for players while they're on the ice) that Vic appears to use once in awhile either. I think Tyler mentioned he was going to broach the subject as well, so I imagine he'll provide some of those numbers.

OK, let's start with Jason Arnott. Over the past five years, Jason has posted 2.36 ESP/hr as compared to Ryan Smyth's 2.31 ESP/hr. I'd call that a pure production saw-off, if there ever was one. On the PP, Arnott has posted 4.27 PPP/hr to Smyth's 4.06 PPP/hr - a slight advantage. Arnott averages about 130min per season less than Smyth with a ES/PP/PK minutes breakdown of 77%/21%/2% as compared to Smyth's 70%/23%/7%. According to Gabriel Desjardins' site, Arnott ranks 102nd in the league in minutes toughness as compared to Smyth's 56th place ranking.

Desjardins also keeps track of penalties drawn and taken while a given layer was on the ice - it's like PIM +/- and probably gives another good barometer of the balance of play. Smyth ranks 9th in league according to Desjardins Not only does Smyth outscore difficult opposition he and his linemates also draw penalties at a much faster rate than any of the comparables listed. My expectation would be that this is common practice for Smyth.

To summarize, Smyth provides the same production as Arnott at ES and draws more penalties all while playing tougher opposition. In fact Arnott may be playing slightly tougher opposition this year because he has played 2nd fiddle for the past few years in Dallas behind Modano, who has soaked up tough minutes for a very long time. On top of that Smyth has been taking on an increased role in the PK unit - from about 3% of TOI a few years ago to 10-12% over the past three years. Is all of this worth an extra $1MM? I think this is a compelling argument on Smyth's behalf.

Sullivan was a great find by Kinger. He is signed for cheap at $3.2MM and his numbers are very good. He's got the edge over Smyth in ESP/hr but has pretty similar PPP/hr rates. The major difference though is the difficulty of opposition. This year, Smyth ranks 56th while Sullivan ranks 89th. That's not a huge difference though and I'm not sure how to quantify the difference in terms of something meaningful anyway. I'd be willing to bet that this is a high ranking for Sullivan this year - the Central division has been terrible for some time and Sullivan has spent the past eight seasons toiling in what has been one of the worst divisions in hockey in almost every one of those years. Besides the question of minutes toughness over the past five years, I'd also argue that Sullivan signed a deal at a low point in the market. He signed his deal in August 2005 shortly after Pronger agreed to a very friendly deal with the Oilers. I think this comparable probably hurts Smyth's case a little bit, but there are some significant holes in it. This is a feather in Poile's cap - even if Sullivan typically plays poorer opposition, he's certainly throttling them and at $3.2MM/yr he's a great deal.

I think Vic said he thought Doan might be the most overpaid player in the league yesterday. I'm submitting Olli Jokinen for that title. Jokinen was in the same situation as Smyth last year and signed a very generous four year contract extension with a cap hit of $5.25MM/yr. In the past five seasons, Smyth has superior ES scoring rates (2.31 ESP/hr vs. 1.94 for Jokinen), the same PP production as measured by PPP/hr (4.1 vs. 4.1 PPP/hr) and Smyth now kills penalties while Jokinen no longer does so. The strength of minutes gap is laughable (Jokinen ranks 404th in the league) and surely a strong argument on Smyth's behalf. I've looked at Jokinen's shift charts in previous years as well and he wasn't spending much time versus the elite players in the East before this year either. Furthermore, Smyth's 9th place ranking in PIM+/- is oceans away from Jokinen's 391st ranking. This is a powerful argument on Smyth's behalf and Jokinen's already up at $5.25MM/yr.

Koivu's next on the list. They're pretty close at ES (Koivu has had some impressive seasons) and Koivu has a decided advantage on the PP of over 1.1 PPP/hr. However, Kinger managed to find a guy that is less durable than the myths say about Ryan Smyth. If you're worried about Smyth at the end of a five year deal, how would you feel about Saku Koivu? On top of that, Koivu doesn't appear to play the same level of competition that Smyth faces, nor does his line draw penalties at the same rate. Koivu's another comparable that helps Smyth's case in my opinion.

I don't have much to say about Doan that Vic didn't already say and I think the numbers speak for themselves. Smyth's just a far superior player. Doan's making $4.5MM/yr and surprisingly is a much better comp than two of Kinger's old men. Doan provides a great argument on Smyth's behalf. No wonder Gretzky was so quick to defend his buddy Lowe after the Smyth deal. Gretz owed him one. Big time. That ridiculous extension threw Lowe under the bus in the Smyth negotiations.

St. Louis is last. Smyth's rates are eerily close to the ones that St. Louis puts up. Again, Smyth blows him away in minutes toughness and drawing penalties and St. Louis averages only about 100 extra min of TOI. Richards' line appears to take the tough minutes in TB and if I recall correctly, he's been doing that for a few years running which leaves the Lecavalier/St. Louis duo to go after the next best opposition. This is another guy signed at over $5MM that Smyth has a clear advantage over.

What else can you say? This was a bad move for Lowe and although we've been promised the other shoe will drop we can't really believe that can we? I have a sneaking suspicion that if the Oilers do get a UFA, they'll make a major mistake on someone obviously not elite like Timonen.

EDIT: The pictures with the data are coming. Blogger's giving me a headache.

36 Comments:

Blogger kinger said...

Riv,

Jokinen? He's 27 years old. You're right about Brind'Amour, the age difference makes him a poor comparison. Despite this fair assertion, you want to compare Smyth to a guy that would be 5 years his junior at the start of their respective deals?

Doan is a fair comparison, but it just shows the other end of the spectrum. Call it a foil to Sullivan's deal if you wish. Also, in Koivu's case, the fact that he's only signed for three years at that price has to play into the equation. I am very supported of 5.5+ over 3 for Smyth; I think the deal breaks when you add on those two last years.

As for the rest of your arguments, I've called all along for someone to quantify the value of playing tough minutes. You can't say your doing a responsible analysis when you ask questions like "[i]s all of this worth an extra $1MM?" without taking the time to actually answer them.

What's compelling about it? How does tough minutes + PK + better at drawing penalties = 1M over 5? I really, really want to see someone show me this.

Do you wanna hash this out on IRC? If I lived in Edmonton, I would prefer to meet up and discuss it. Unfortunately, I call Vancouver my home, and discussing these sort of things over bloggers chat feature is pretty pointless.

- kinger

3/03/2007 4:32 pm  
Blogger kinger said...

"I am very support[ive] of 5.5+ over 3 for Smyth"

<3 proof reading

3/03/2007 4:34 pm  
Blogger RiversQ said...

kinger said... You can't say your doing a responsible analysis when you ask questions like "[i]s all of this worth an extra $1MM?" without taking the time to actually answer them.

Hey ditto to you buddy. What did you say?

Here is where you make up the ground between a Jason Arnott and a Ryan Smyth. Their numbers may be similar, but Smyth has taken a much harder road (especially in the last couple years). Still, is this ability really worth in excess of 1m dollars per season on top of the fact that Arnott outscores Smyth? I just can't bring myself to believe that it is.

That's the same thing.

Of course the last part is pretty much wrong. Arnott doesn't outscore Smyth on a per hour basis and furthermore he doesn't even play more minutes. Smyth has nothing but advantages in that comparison.

And what about St. Louis? Same rates, tougher minutes for Smyth and he's already getting $5.25MM/yr.

With the Jokinen comp as I was trying to get at the idea of elite players. Jokinen is an example of a player people hold as elite. Smyth has matched his numbers while playing much much tougher opposition.

Koivu's probably just a bad comparable all the way around.

I've got other comps as well that I didn't bother to put in here. Savard, Tanguay and Gagne are three more players making $5MM+ with numbers that approximate Smyth's while playing easier minutes. Younger players? Sure, but still not as good as Smyth has been and probably not as good as he will be over the next 3 years.

Basically, Smyth sits smack dab in $5.5MM/yr territory in terms of results and I have not taken any of his other more abstract attributes into consideration.

3/03/2007 5:36 pm  
Blogger RiversQ said...

kinger said...

"I am very support[ive] of 5.5+ over 3 for Smyth"


Hey, who the fuck isn't?

The problem being that the Oilers aren't signing a UFA this offseason that's as good or better than Smyth that's going to agree to a 3-yr deal.

If the Oilers want to win at any point in the next decade they're going to have to pay market-rate for some players. Until they decide to do so, they will never be competitive.

3/03/2007 5:39 pm  
Blogger kinger said...

Blogger comment arguments = the suck.

MSN? IRC? GTALK?

Quick comments:

- It's your assertion that Smyth is worth more than Arnott because of his ability to draw penalties and the fact that he plays against a higher quality of competition. It's on you to prove what you say to be true before it's on me to prove that it's false.

- St. Louis, PK+/-? He absolutely destroys Smyth in this department.

St Louis: -2.07
Smyth: -4.14

- I was referring to Smyth and Arnott over the course of their careers, not just the current season (is this not a prudent thing to do when handing out long term contracts?).

3/03/2007 5:48 pm  
Blogger kinger said...

The problem being that the Oilers aren't signing a UFA this offseason that's as good or better than Smyth that's going to agree to a 3-yr deal.

Foresight is 20/20?

3/03/2007 5:50 pm  
Blogger RiversQ said...

I was referring to Smyth and Arnott over the course of their careers, not just the current season (is this not a prudent thing to do when handing out long term contracts?).

I am referring to their numbers over the past five seasons, which is far more relevant to the proceedings than their career numbers. Arnott has not been a superior producer. In fact Smyth has put up 19 more points in 15 fewer GP over this period of time.

I'll take a kick at the PIM+/- thing.

The league average PPGD/hr rate last year was 6.0 PPGD/hr. Assuming all these guys play their average year and have this year's PIM+/-, and average PP units playing for and against them here are their GD contributions based on PIM+/- alone:

Smyth +5
Arnott +3
Sullivan +3
Jokinen -1
Koivu +2
Doan +4
St. Louis +4

So yeah, unless I've missed something it's not a huge difference.

St. Louis is playing for the 28th ranked PK unit in the league and he's driving his PK +/- with SHGF. SHGFs are probably one of the least sustainable numbers in the league outside of possibly ENers.

3/03/2007 6:09 pm  
Blogger Slipper said...

Just because he's on the penalty kill doesn't necessarily mean he's successful there. By Dejardin's numbers (If I'm reading them right. There must be an easier way to express those integers) it looks as though Smyth is posting one of the league worst Opponent +/- ratings while short handed, and the second worst over +/- for forwards who have played over one hundred minutes on the PK. BTW, are Pisani and Reasoner ever fucking killer by Dejardin's metric! Now those two would have one hell of an argument to get a little payola for there short handed efforts.

I'm not really fan of this trade because of the lie of a season and the wasted opportunities it exposes. Also because the player probably could have been had for 4 at 4.8 last off season. Lowe gambled with my love when he entered the season with an unsigned Ryan Smyth and a blueline that could barely match-up to the AHL affiliates (oh man, let's not get started on that) of the other team's in the Northwest division.

With the current NHL schedule a team's not going to compete if they aren't atleast up there with the cream of their own division. The Oiler's aren't even close. Roli can't steal games like Kipper and Luongo, the defence is 3.5 veterans short form the cup run, while Smid (who I think will be a player) and the 4 defensmen the Oilers are going to need to acquire at some point probably won't be ready for prime time until 2010-11. There's also atleast two coaches in the northwest tht appear to be able to squeeze blood from a stone. I couldn't say the same for MacT. He could reign in the butternut squash line and spread out the good players on the team in the early part of the season, and I think that helped in the top ten draft pick drive.

Not a year away, or even two. With or without Ryan Smyth. And barring some miracle in the off season, I can't see how any marquee player could change that. Spending 5.5 to 6 on any player in a rebuild year is pretty pointless. You should either challenge for a good shot in the post season or a good pick, and Smyth can tip the scales enough that a team heading toward the latter finds themselves somewhere in the middle.

That in no way absolves Lowe from creating the mess in the first place. Sending 94 to Long Island AT THIS POINT may have been the right decision, but one arrived at as a result of all of Lowe's other failures.

3/03/2007 6:11 pm  
Blogger Slipper said...

Kinger. Better opposition play more, and in effect have a larger impact on the result of a game.

3/03/2007 6:15 pm  
Blogger kinger said...

Kinger. Better opposition play more, and in effect have a larger impact on the result of a game.

Clearly...

3/03/2007 6:20 pm  
Blogger Lowetide said...

The single strongest argument for Ryan Smyth is "established level of ability." He's been a strong, strong player against the middle of the order for some time and that's the meat of the issue. We're talking results over 80 games when the evidence is 250+ strong.

3/03/2007 6:25 pm  
Blogger Slipper said...

... Well, if you're logging more minutes, and they're against the players who have the biggest impact on the scoreboard, and in situations like PK, PP, and zone draws which all tend to greatly affect the game outcome, all the while generating positive results; isn't that contributing more to winning?

Aren't players being paid for their contribution to winning hockey games, and not their placement in the scoring race?

3/03/2007 6:45 pm  
Blogger The Rage said...

Essentially, a player's worth is determined by the contribution to team goal differential, right? So wouldn't the best way to evaluate a player be to compare team goal differential with the player in the line-up, to the differential with the player out with injury? You would probably need multiple seasons to get an adequate sample size, but I think this method would show a players worth to a particular team very well.

Also, to compare the ES abilities of players who play different toughness of minutes, could one simply add opponents ES GD (goal differential) with a players ES GD? In other words, if player X plays opponents that have an average of +1 GD/hour, and he is +0.5 GD/hour while doing so, then isn't his ES worth +1.5 GD/hour? That player prevents 1 goal differential per hour, and creates 0.5 goal differential per hour.

3/03/2007 9:48 pm  
Blogger Chris said...

Geez, do we really need to go into a bunch of math to point out that Jerome Ignla, Joe Sakic, Mike Modano and Peter Forsberg he's not?

3/04/2007 12:07 am  
Blogger YKOil said...

piece if farkin' c-r-a-p.

sigh

lost my post when I had to sign in using the google account.

point was that Smyth has only even played with a decent defense once - last year. he also survived salo and Conkannen.

For Smyth to provide the numbers he has - as long as he has - is freakin' amazing. A lot of HF guys say Smyth isn't elite.

This is the first year in a loooong time Sakic has played in front of a horrible goalie, with a porous defense and without a great 2nd string guy to back him up (Forsberg). His year is no great shakes.

Point being - give Smyth an 8 or 9 year stretch playing with guys like Forsberg, Blake, Bourque, Roy, etc and we wouldn't be having this conversation about whether or not he is elite.

3/04/2007 1:16 am  
Blogger YKOil said...

oh hey - I have a blog now btw

ykoil.blogspot.com

3/04/2007 1:18 am  
Blogger PDO said...

Rage, I'm all on board with your second idea there, but there would never be a significant enough sample size to determine the first part, especially given the fact that over the course of the seasons the player would be improving or deteriorating.

3/04/2007 10:15 am  
Blogger Julian said...

Surely Smyth contributes to a teams powerplay success in ways that don't show up in the PPP/HR rate, if you agree that he does as good if not a better job than anyone at annoying and screening the goalie.

3/04/2007 11:23 am  
Blogger joninabox said...

Rivers, I think one of the main problems with the whole Smyth debate is that no one is ever arguing about the same points. :)
You guys probably know where I stand on the Smyth issue if you've kept up with the BoA thread, but I have to say, your post here does reasonably well to show that Smyth is probably a 5.5 million dollar player right now (as a sidenote, I should just point out that ALL of those comparables make less than Smyth was offered). The problem is, I don't disagree that Smyth is a 5.5 million dollar player right now. In fact, he's probably a 5.5 million dollar player for this year, the next, and possibly the following year. But the last two years of the contract? It's obviously impossible to know how Smyth will decline in his twilight, gradually or dramatically, but as a GM Lowe has to look at precedent and history and assume that Smyth is not going to produce anything like a 5.5 million dollar player at 35 and 36. Hell, we have a guy on the bench "coaching" who's a pretty damn good example of the toll age can take on a player who scores a lot from in front. Given that it's not a good bet Smyth will age well and that most of our young core will be peaking two or three seasons from now, it's really hard to give the OK for a 5 year contract at 5.5.

3/04/2007 1:25 pm  
Blogger Asiaoil said...

That is the issue in a nutshell - not that Smyth will not perform well in the next 2-3 years - but what will he contribute in Year 4 and 5. The Smyth boosters are largely using this year as the baseline and just assuming this is what we will get until 2012 - and that is foolishness. Sure Smyth could be worth $5.5 million for 3 years on a team with another young undepaid dominant player that he could complement (if I was PIT I would have chased Smyth HARD). The problem is - the Oilers are rebuilding the next couple of years and we also don't have another young dominant player for Smyth to compliment (Hemsky is also complimentary but unlike Smyth he is signed longterm to a nice contract and will improve). Smyth is a winger and he just can't take the team on his back like a dman (Pronger) or a goalie (Luongo) or even a dominant center - why is this so hard to accept? He just doesnt play the minutes those guys do or have the puck on his stick enough. It's also getting irritating Riv when you toss out any good comparables for Smyth (like Sullivan) and focus on the bad contracts - are you his friggin agent or something? The fact that other GMs are stupid is not a reason to throw open the bank vault for Smyth - only agents spew that crap as rational thinking in terms of building a team.

Smyth over the past few years at $3.5 million was a great moneyball player - but we are not dealing with the past except in Riv's analysis. At $5.5 million over the next 5 years Smyth is a good older player who is will likely deliver fair value for the next couple of years - but is a BIG risk after that. How many of these UFA deals have we seen work out for the 31 and over crowd over the years? - damn few - and only in situations like DET where money didnt matter. Look at the Weight contract for a close to home example of money poorly spent on an older UFA. Well money does matter in a capped system and spending huge amounts of it on vets in their mid 30s with declining skills is just bad judgement unless you are the Yankees and dont give a shit about cost (not that this is working out well for them either).

What I want is to build from goal out - not load up on expensive complimentary wingers - we have enough of those. Roloson is OK for 2 more years - but we need a dman to go with the young core. JayBo is the man - young and already surprisingly decent at ES. Redden would not be bad either and it would take less to get him than many think if we were eating a lot of salary - but JayBo would be better. If DD or JDD are ready in 2 years then we have a cheap goalie to go with our dominant dman and young forwards hitting their prime. If one or two of the remaining prospects to pan out we are set pretty nicely for another cup run where we load up with soon to be UFA vets.

Until the summer I am willing to give Lowe the benefit of the doubt because his record is pretty damn good in terms of getting rid of guys at peak value right before they drop off. The Smyth frenzy is SO similar to the Niniimaa deal it's laughable (tears and all) and that turned out pretty well for us - plus Lowe's move's last year when we had a legit shot were solid if slow - but again I'm willing to believe that Roloson was the guy he wanted and he waited until he got him.

At the end of this season the ballsy and correct move for a GM would have been a complete rebuild (and I posted as much last summer). Pronger split and we shot our load with solid rentals like Peca, Spacek, Samsonov, Tarnstrom - close but no cigar. The makeup of the team makes that the only sensible move. Well we are clearly rebuilding now even if Lowe did not say as much this year until now - his actions (or lack of them) said everything and he traded Smyth at his peak value point. I expect serious moves this summer and even if Riv and Dennis don't think it will happen - recent history (Pronger ) says it is possible and we will see how things evolve this summer especially with teams like OTT, TBL needing cap space and FLA needing to rethink their direction.

Asia

3/04/2007 3:20 pm  
Blogger Vic Ferrari said...

The thing Rivers, is that from a purely business point of view Smyth is worth $8 million per. And a lot of fuckers know it. And he's not coming back. Because he'll sign for less somewhere. Charles Wang is treating his family like rock stars for Chrissakes ... while Kevin Lowe is on television educating "mo" and his buddies that Smyth wasn't elite ... Charlie will be flying out Stacey and her kids on a private jet with a children's entertainer on board. And just wait until the some of the other teams start schmoozing on July 1st ... dude's ego will be through the roof and I doubt he will be coming back to the GM who shat on him to protect his own.

Did you catch any of the Comcast broadcast last night, rq? When that pretty little thing jumped on the Caps bench and asked the Caps coach (Bob Johnson lookalike) about the game plan.

It's pure gold, start to finish. Including what had happened on the ice for the minute previous and then the minute prior. Post it on IOF if you have it, just killer stuff.

3/04/2007 3:28 pm  
Blogger kinger said...

Hey Riv, I did a piece on Timonen on my site. You should check it out, he really looks like a guy we should be going after this offseason.

3/04/2007 4:43 pm  
Blogger Asiaoil said...

Vic if Smyth is so fucking good and worth so much - why is he a fringe Team Canada player on par with Doan? - why have we been so bad with this monster on our team over the years? - why did it take Pronger, Peca, Roloson, and Pisani morphing into Rocket Richard for 2 months plus about 5 other solid ES vets (Horcoff, Moreau, Smith, Staios, Spacek) for us to do anything last season in the playoffs? You guys are shuffling numbers to make the guy into Gordie Howe - and he just isnt. Good player no doubt - but spending that kind of money on a late career complimentary guy like Smyth while you are rebuilding makes ZERO sense. A lot of very smart hockey people can see that too - not just internet hacks like me.

You seem to think he will be like Shanahan in his golden years - problem is - he never was as good as Shanahan in his prime so why would he elevate to that level later on. Kinda like a decent looking 40 year old women wanting to look like the older Hollywood babes - probelm is if you never looked like Liz Hurley at 20 you sure as hell won't look like her at 40. Smyth at $5.5 is fair value for for 2-3 years on a team with young, elite, under-priced talent looking to make a run - the Pens should grab him this summer to get the most out of Crosby now. On a rebuilding Oilers for 5 years I have HUGE doubts. You also just can't assume what you get this year is what you will get 3-5 years from now - and going on about performance in his 20s will not change that one little bit. He was solid value during the cup run with the team as structured last year - this season it made sense to sell him at peak value and move on.

3/04/2007 5:15 pm  
Blogger Vic Ferrari said...

Asiaoil:

I'm nothing but consistent. When you and others like you were embarrassing yourselves by writing Pronger homoerotica I was trying to tell you that Smyth was the only .25 goal diff player on the team. This was during the playoffs, look it up. I was as right then as I am now, and I take no joy in that.

I've only ever given hockey pool advice once ever before as well. That was around Christmas when I advised everyone to nab Zetterberg, who was having a disappointing season at the time but had underlying numbers that kicked ass. Again, look it up.

Now will be my second endeavour at hockey pool advice. Take Satan.

Why? Not because he'll probably play with Smyth, but because he'll probably play the shift after. The opposition is going to start seeming a little slower for the l'il gypsy ... and that's because they will be.

You lads love Hemsky, bless ya. And he's a good player. But at the very prime of his career as a difference maker (32 years old or so) it is very likely that he will have anything close to the impact that Smyth does on results. I'm sorry but that's just true.

I feel pushed here. You and Sacamano, I wouldn't say you're being bullies. But man. Fuck it, I'll put my money down, we'll drop the Oiler lines a dime for Smyth's departure (TWICE Hemsky's value btw) and increase the Isles chances just the same for the Isle games. I'll take the Oilers and Isles on all counts. It seems childish, and I shouldn't bite I know, but youn know AO that I'm not one to back down, and like sac says it's the fair way. Even if it's only 5 or 10 thousand a game, at least it keeps us paying attention and having fun.

3/04/2007 7:05 pm  
Blogger Asiaoil said...

You seem to care about what's happening this year - but I'm not arguing Smyth's value right now - the Oil are set up to do something 3 years from now and Smyth's contribution next year won't mean a f'ing thing in the grand scheme of things. Let's see what the boy does in 09-10 when he's starting to get old, fat and happy to just punch the clock for a couple more years before sailing off to cliche heaven......

3/04/2007 7:27 pm  
Blogger Vic Ferrari said...

AO and sac:

Okay something on the sidebar now. We're ready to go. Just a little over 3 point holds there, so that should be livable for all of us.

I'll give the the straight lines for the most recent games against the same opposition in the case of both the Oil and Isles. I'll use 6AM lines if that's acceptable. I'll give you a tenth of a goal diff for the Oil and you take back the same from the Islanders.

This should be straight forward, there are a couple of ways to make sure the cash is going through. If either of you guys want to run this another way I'm totally flexible.

3/04/2007 7:42 pm  
Blogger Slipper said...

Are you serious ferrari? 5/10K per game?

You can't possibley expect anyone who trolls Oilers chat logs with the wealth of their time to cover that kind of action. If you can, with your stats and skills and ability to analyze it, you should carry your own book.

3/04/2007 8:00 pm  
Blogger Vic Ferrari said...

5/10k is nothing. You know better than this. For crying out loud a monkey betting randomly could wager a million and only be down 30k at the end of it. And that's a monkey ... what about a smart hockey fan? It seems like a lot but it really isn't, and for Christ's sakes it's not like any of us are working at a carwash. Thanks for your contribution though, slipper, that really helps. (You are "biff", no?)

3/04/2007 8:30 pm  
Blogger Asiaoil said...

Vic - gambler bluster means little to me my friend.......I'm a patient man and I'm happy to wait 3 years to see if I'm right and you and Riv are surviving on your island drinking each other's piss. Another little point that most stats guys or modelers know all too well but never say out loud - you can say think mathmatically that would get you laughed out of the room if you said them in common English. If I heard right - you basically said Smyth is more valuable than arguably the best dman in the game - come on. Well Pisani scored like Bosy in the same saids playoffs so he should be in the HoF - right?. Nobody is saying that Smyth is not a good player - he makes a difference most nights now in the heart of his career. I just doubt that he makes $6 million worth of difference in his mid-30th and there's nothing astounding or fantastic about that particular thought.

3/04/2007 10:13 pm  
Blogger Slipper said...

Hockey's the easier sport to pick compared to the points, but the line puts people to the greater test. My betting experience is limited to only one book and the Canadian lottery, so I'm still in my infancy when it comes to this shit. My book is ten percent juice on the points (the juice never moves but the points do), so I figured I'd have to show myself that I could carry well over 53% before I'd ever allow myself to bet big bank. Because at that number you're just carrying the day.

Hockey makes you put down the money to get the good wagers though. I got killed earlier this season on a PHI@ANA and since then have experimented in only seeking the undervalued faves or the good dogs, because the book takes the dogs straight up.

More to the point though, I think you might be taking the whole debacle to personally, for lack of better word. I've try to communicate the things I've learned both from browsing these math geek sites and from betting to everyone I know or come across and the majority of the response is blank stares. Three days before the Bergeron trade I was trying to explain the linemate quality and opposition and all that shit to this guy; it was a diatribe with me clumsily trying to tell someone why what they thought what they were seeing wasn't really what they were seeing. People aren't really warm to that idea. All I got back was a blank stare and "I hear what you're saying, I really do.. but I think we really just should trade Bergeron". It'd be frustrating if it weren't so fucking funny.

I thought that maybe it was only people that bet that could get it. If you're trying to profit off of runs and baskets and goals, surely you'd eventually be drawn to the value of these points, no? No. I had a kind of crew of pals, and we were all betting together, and they were all gung ho about analysing the socres and not spreading your money thin, and blah, blah... Then as soon as an early game we wagered on went into the shitter, they'd literally run to one; they'd be on the phone in the bathroom stall making blind wagers on the late games to try to recoup their losses. My friends would call me and literally say shit like "X team has lost X ammount of games in a row, they're DUE", as though somehow losing games increased your odds of winning one.

It's no fucking mystery to me anymore how people lose so much money on the shit.

In fact, the only people whoever listen and buy into alot of the ideas I steal from around here are my nerd friends who barely watch a lick of sports, if any at all. ALl they'r einterested in are data, so somehow it fits right inot their paradigm. That's just the reality.

3/04/2007 10:38 pm  
Blogger RiversQ said...

Asiaoil said...
Vic - gambler bluster means little to me my friend.......I'm a patient man and I'm happy to wait 3 years to see if I'm right and you and Riv are surviving on your island drinking each other's piss.


Now, now, you're already on the record as bailing out if they don't spend that money this offseason. You probably won't be around in three years.

I just doubt that he makes $6 million worth of difference in his mid-30th and there's nothing astounding or fantastic about that particular thought.
Yeah we'll see I guess. Would you know the value if you saw it though? I'm not convinced.

3/04/2007 10:48 pm  
Blogger Jeff J said...

Slipper said:
"It's no fucking mystery to me anymore how people lose so much money on the shit."

Interestingly enough, those people are the ones indirectly setting the lines. The bookies don't have some magical formula saying Smyth is worth 0.25 goal diff per game. They just set the lines so they have roughly the same amount riding on each side. It's the bettors who collectively "think" Smyth is worth +0.25.

Personally, I haven't got the faintest clue how much of a real goal differential Smyth contributes versus Pronger. I would caution against evaluating the worth of a player according to how much the betting masses think he's worth.

3/05/2007 12:37 pm  
Blogger Vic Ferrari said...

jeff j

That's completely wrong. Any bookie may drift the lines a bit later in the day if they get a bit too much action on one side of a wager, depends how much risk they are willing to absorb. On all the lines they post, on any given night, they'll lose on a lot, win on a lot, and end up tweenerish (+/- 3%) on a handful.

In hockey the holds are around 3.5% and the average profit should be around 6%, or close do it. The gaming commission in Nevada publishes this stuff annually, if you really care you can dig it up.

I mean if you get a stretch of NFL games where the favourites cover the spread, then the bookmakers, collectively, well they are going to lose millions upon millions of dollars. In the end they'll get you though, and for more than the hold.

The lines should be set to maximize profit. You are betting against the bookmaker, NOT the market.

3/05/2007 3:45 pm  
Blogger Vic Ferrari said...

slipper:

You know I remember you saying you were working as a bartender and I just assumed you were running a quiet book.

Not sure what you mean about the underdog though. Check the rules for each place, but if the underdog is +135 then you'll get your wager plus 135 bills back if you win.

And I know what you mean about guys chasing losses. In college football I've sometimes wondered how effed up the odds are for the late start college games on the west coast. Or especially the U of Hawaii games. I'm thinking that if you always took the heavy favourites (in games where there was one) that you'd end up beating the line in the long run. I'd think that the bookies usually make a killing on those games, but I dunno.

We don't get any payment for links to our advertiser here btw. Just a flat rate for the year.

And Sagarin is actually very good. Use his predictor numbers on the right and he's ever so close, usually within a couple % of actual odds. I think the guy uses pure math though, so if Zetterberg gets hit by a bus on gameday, or Sutter decides to start MacLennan ... it's not going to affect his values. It will eventually as he weighs in recent games, but not on the day I don't think.

Maybe I'll write a script to see how Sagarin would do over the course of 100 games or so. Dolphin as well. Just for the helluvit.

3/05/2007 5:06 pm  
Blogger Jeff J said...

"The lines should be set to maximize profit."

OK Vic, it may not be half each side - fair enough. But the bookies still only set the lines in reaction to the market. When they say Cowboys are -3.5 favourites over the Redskins, they are not saying that over a million samples the Cowboys will average 3.5 points better than the Redskins per game. As you said, the bookies figure that setting Cowboys by 3.5 will maximize their profits.

If Michael Irvin gets arrested just before they start taking bets and the spread swings to the Redskins by 3.5, the bookies are not saying that Michael Irvin is worth seven points per game. They are saying that a sufficient number of bettors think he is worth seven points per game and the new line is set, as you said, to maximize profit. They are still only reacting to the market.

3/06/2007 6:14 am  
Blogger Vic Ferrari said...

Impact of individual players on game results is tricky stuff. Of course the market has an impact, and every bookie has an acceptable limit of exposure. That's their decision.

So if slipper is down at the bar, and the Redskins are playing vs the Cowboys in the Monday night game that evening, and he has good reason to believe that the Cowboys only have a 47% chance of covering a 3.5 point spread (an exaggeration for effect here):

Then if Slipper is running -110/-110 with the Cowboys as 3.5 point favourites, and if the customers pony up a total of $1500 on Dallas and $500 on D.C.. Then arithmetic tells us that he'll either lose $865 or win $1045. Now if slipper had a good Sunday he's probably happy with that. If he's correct on the true odds then he's looking at a good hold, around 7 or 8% or so. Well worth being in business at that clip.

If he's had a bad Sunday, plus he's got a car payment and taxes due the next day ... maybe he gives a last minute deal on D.C, moves the spread up or goes -120/100 .. just to draw some more action to the Skins. And if even that isn't enough then maybe he takes a few hundred and bets it on the Skins through another bookie. He's losing potential profit with every hedge. He's got to balance risk with margin, that's entirely slipper's business.

3/06/2007 11:24 am  

Post a Comment

<< Home