|
RotoGuru Menu Football
Baseball
Basketball
General
|
|
![]() Feedback This page is devoted to feedback I've received from you. If you have a question to ask or a comment to make, send me an email and if I think it's of general interest, I'll post it here.Click here to move ahead to current feedback letters. Guru: Still some life in the Deadskins, eh? - Guru Guru: That leads me to Miami. I probably will take Atlanta's defense over the NY Giants, because I think Miami is breathing a sigh of relief that they are in the playoffs and hope the Jets do the dirty work for them. Plus, other than the Oakland game, Miami's defense has never played on the road the way it has at home. I know the game does not mean anything to Atlanta, but it's been so long since they've been a winner, I don't think they want to go into the playoffs on a losing note. Also, if history continues, the NY Giants game should mean something. Tampa Bay has never won a game when the temperature was below 40 degrees. They couldn't even win a must win game last week when it was in the 40's. As far as Philly goes, even though they are only 1-3 since Detmer has started, he has only been sacked twice (none in the last 3 games), and only thrown 3 interceptions. Even though Philly only scored 9 points last week, Dallas had a -401 SWP. So Philly has been getting some yards and Detmer has not been turning the ball over. Who knows? Another major factor is, I already have Atlanta and I can talk myself out of just about all the other possibilities, so if I keep them, and I can make one more wild trade for the last week. - Steve H. (12/23) response from Guru:Good observations. Of course, when it comes to team defenses this year, it seems that history isn't the best guide - at least for the teams I've been picking. And just because TB can't win in cold weather doesn't mean they're not a good defense to pick. Even losing to Washington last week, they posted a very respectable -286 SWP. - Guru Guru: I did consider Testaverde for one of the poll choices, but went with Young, Cunningham, and Favre on the basis of their higher SWP totals. Plus, I figured that if I included Vinny, I'd also have to include Flutie, since their stories are very similar. And the Giants could be the best pick, especially if Philly mails it in this week. On the other hand, if Tampa Bay wins the early game, then the Giants will be emotionally down. Tough call. - Guru Guru: Could be. The Ravens were pathetic against the Bears last week, too. As an ex-Browns fan, it's nice to see Art Modell getting his just desserts. The Browns just might finish ahead of the Ravens next year! - Guru Guru: I think if SW had just stuck with their preseason plan of 50 trades max for the season, and $1,000,000 max weekly change, that inflation would have been much more under control. I don't understand why they changed these rules, especially after the season started. But given the extra volatility and the extra trades, I think the end result was quite predictable. - Guru Guru: My guess is somewhere in the 40,000-50,000 range. Check the RotoGuru team rankings. Teams with rankings in the 40,000's still appear to be giving it a go. I also know of one team that hasn't had any trades since opening day, and is ranked 57,000. So I figure the answer is somewhere in between. These numbers relate strictly to the Smallworld site. At the CNN/SI site, the comparable cutoff seems to be around 70,000. - Guru Guru: I don't think adding extra players really addresses the issues you are complaining about. By and large, the "problems" you cite are related to a short schedule and four trades per week. The only sure-fire way to reduce the amount of "game picking" is to reduce the number of trades. And, SW actually increased the number of trades after the season started. While this certainly leads to a lot of swapping from game to game, this is the only activity managers can do during the season, so reducing the trade allocation may not be in the best interests of maintaining manager interest throughout the season. - Guru Guru: Jamal Anderson will be my odd man out because Mike Ditka might finally have the Saints defense playing a little defense, and they have a QB (Collins) that can reach a WR now and unbelievably in the NFC the Saints (argh) are in playoff contention. I don't know if they can win, but it should be a tough game. That said, Anderson will probably have his best day, and Taylor his worst. My hunches have probably not been as good as Eric's. - Steve H. (12/11) response from Guru:We'll see soon enough. - Guru Guru: I agree with your rationale, and in fact, that's why I didn't pick up Taylor. No shortage of good opportunities at RB, Brunell's absence makes the Jaguar's offense more speculative, and the money is irrelevant at this point. That said, SW managers, in general, chase after hot players, particularly those who just dropped the maximum in price. And not all SW managers are financially gorged. So while there are valid reasons not to trade for Taylor, I'm particularly surprised if the majority of SW managers are following such a thoughtful line of reasoning. - Guru Guru:
I have had a very productive past two weeks with one of my SW football
teams. Unfortunately, it was with the worst of my two teams at the time
-- isn't that how it always works out? Anyway, I managed to rack up
5988 points in a two-week period (I was hoping to break 6000, but oh
well). You may have noticed (probably not, but I did) that my team
(Fightin Ags) was fourth last week and third this week in Gurupie
rankings by total points per week. What's more, I did this without
making a trade in that two-week period. That got me to thinking, "What
was the best possible team to have over that two-week period and not
have to make a trade?" So I did some calculations (partially inspired
by your daily blurb today) and figured it out. Here it is: These calculations were done rather hastily, so it is possible that something is wrong, but this should at least be close to correct. Now my team matched on some of these, but not all of them by any means. I was disappointed my score wasn't closer to the maximum for this two-week period, but I guess I'll be happy with what I got :). Why did I write this you ask? One, it's dead week here at college, I'm bored, and I don't want to study for finals yet. Two, I was curious if any other Gurupies had a two-week total that broke 6000 (and if so, can you say you did it without using a trade?). I'm sticking with the same exact team for this week again, too, so we'll see how it goes. Thanks for just letting me ramble on. You probably won't even post this. - Matt S. (12/9) response from Guru:I'm impressed. I haven't had a 3000 point week once, and you have them back-to-back without a trade. If I had Miami the past two weeks, I'd have been close. In fact, the lineup I had last weekend, with Miami instead of Minnesota, has produced 5804 SWP in two weeks. Unfortunately, I didn't have Miami, and I didn't have it for two weeks. It's a game of inches. - Guru Guru: In fact, price changes are totally dependent on buys and sells. Performance has no impact - other than that good performance tends to lead to buys, and vice versa. Is the price change formula a set amount per buy/sell? I don't know. Could be, but Smallworld doesn't publish the details. I do know that the initial drafting of players has no price impact, only subsequent trading activity. - Guru Guru:
I agree on most of your points. But I don't think your concern about a handful of managers getting penalized for buying someone at too rich a price is warranted - as long as it's stated up front that this is a consideration in pricing. And I do think 64 trades is a lot of shuffling for 17 weeks. I don't think I ever found myself wishing I could make a trade this year, but not having the trade available to spend. Takes away one of the strategic challenges. - Guru Guru: There are more effective ways to solve the roster inflation problem. Seem to me like picking special teams from week to week would be a real crap shoot - sort of like picking a kicker. I don't think it would be a useful addition to the game. - Guru Guru: Getting into the defense discussion, I have to admit that I've been absolutely socked by this position. New England's defense has been a constant disappointment whenever I've taken a chance on them. I've been pretty good about getting good values out of most of the other positions most weeks, with exception of Barry Sanders recently, and Marvin Harrison's injury week. I'd like to see less emphasis on the "hot player/team of the week", and more on having a consistently good team. I think this would be alleviated by lowering the impact of a poor defensive selection. I also like the idea of giving positive points for defensive scores. Penalizing a defense for its' offense's miscues is a poor idea. Win/loss points are also a good idea. This allows teams like Denver, for example, to look a little better when they spend the second half in prevent mode, allowing a few points while preserving the "W". This version of SW Sports has differed from the "build values, then get points" versions that I was familiar with. I had assumed that football was going to be similar, forgetting the length of schedule affected this. I have to say, it's great for those players who get in late, but I'd like to see a game more like past SW sports versions. I thought that putting players into automatic divisions was the way for players who joined late to still compete. Lastly, I'd like to suggest a modified "trade governing" system. I like the regulation of trades, but perhaps allowing a few more early, and less later would be a better way. For example, allowing a user to make 5 or 6 trades in the early weeks and only 2 or 3 later weeks would be a good compromise. This would allow those users who choose to save their trades, while also allowing users to trade slightly more to generate a higher roster value. - Jeremy B. (12/7) response from Guru:The San Francisco defense phemonenon is one that I've addressed on other sports as well. When players are overpriced at the beginning, of the season, no one buys them in the initial draft, and hence, there's no way for their prices to come down to a reasonable level. I've suggested in the past that there should be some drift mechanism in the pricing that provides for a decline in the price of overpriced, unowned players - but SW has never seen fit to implement this sort of thing. I don't think I follow your train of thought as to why reducing the impact of the defense slot would lead to less emphasis on the "hot player/team" of the week. Reducing the number of trades would have a better likelihood of accomplishing this, I think. While distinguishing between points scored by the offense vs. the defense seems like a good idea on the merits, I think the problem is one of administration. As far as I know, there aren't good sources of data that distinguish between points scored by the offense vs. those scored by the defense. And I think that had SW stuck with it's original trade allocation plan - 4 trades per week, subject to a max of 50 for the season, then we'd have had the effect you are looking for. When they eliminated the max of 50 for the season, it effectively eliminated any need for trade conservation. Four per week over the whole season is just too many trades. Why they changed that rule after the season started, I'll never understand - but I think it materially diminished the need for thinking ahead. - Guru Guru: In last year's SW Hoops, they updated prices every day, which, with only 50 trades for a season, meant a lot less trading for hot players, especially in an 82 game schedule. If a player scored 40 points in a game, most people wouldn't use a trade to buy that player, unless they really hopped up the top 50 list. Many trades had to be saved for injuries. In football, when there are only 16 games and a total of 64 trades, injuries are not even that much of a concern as far as conserving trades go. A player who gets injured is only one of four possible trades the user has to make. This leads to another point, about the 4 trades a week rule. This helps a lot of people as far as trading goes, because instead of making 5 or 6 trades a week, you are limited to four. This means there is nobody left in the cold at the end of the season, who used up their trades too fast. In baseball this year, I also had a roster over $100 million, but I was barely able to get all the good players. I still had about 3 or 4 slots I wanted to change, but could not because of lack of trades (that I had used to build up the roster value), and because of the money constraint. The lack of huge price changes in baseball also caused the roster values to be much lower. For a player to have a good week and considered for buying before the price update, a player had to do well in about 6 or 7 games in a row. In football, just one big game causes big buying, and one weak game causes big selling. This especially works in favor to people who have players like Randy Moss, since a big game will cause his price to go up, but a weak game will cause his price to go down again, at which point people can now buy him cheap, with a few million dollars in cash bonus for selling him. No other sport had this type of trading. Players can have one good week and hop into the top 5 or 10 of their position. In baseball, over a 162 game season, it was more consistent and the top 5 were usually the top 5 for most of the season, which meant their prices were always high. In football, players like Eric Moulds can have two big games, and because of his cheap price would cause major buying. This is the same case as that of defenses. The cheap defenses are often bought simply because they are playing a weak opponent. The San Francisco defense price of over 9 million vs. a team like Green Bay sounds ridiculous if you can get a cheap defense playing Philadelphia. Also, there are only 11 spots in football including two positions where a good player may be very cheap, tight ends and kickers. Shannon Sharpe may be the only exception, but the #2 and #3 TE's over the season have usually been quite cheap. The kickers are all basically cheap, and since they aren't that big a deal, you never know if a 2 million dollar kicker might get more than a 4 million dollar one, even if it's only a bit better. It's quite odd, a friend of mine in my division said to me how he was upset how the top 3 teams in our division all have the same team, except a few players, and how having a big roster value wasn't so much an advantage anymore. I then told him I thought the same thing about him, since he had basically the same team as me. - Sean C. (12/5) response from Guru:The best strategy in past SW games has been to focus on building value early, get the best players possible, and as long as you assembled your dream team soon enough, just allow it to propel you past the competition. This worked beautifully for me in hoops, and even in baseball - although a little more emphasis on point generation early in the baseball season would have served me well. In football, the problem with this strategy is threefold:
So, I haven't yet had the gumption to drop Young and Davis, in spite of the obvious intellectual rationale for doing so. Instead, I've tried to focus on team defense, since that slot has a disproportionately high scoring impact, and seems to lack a widespread consensus from week to week. But it's certainly a different game than I've become accustomed to. Not that there's anything wrong with that. . . . - Guru Guru:
If you maintained this average each week, you'd have a little over 25,000 points. Quarterbacks: This shows that with Steve Young as QB#1, you still have plenty of choices for a QB#2 to beat the average. This looks easy. Even room for a few bad weeks, or injuries. This should be easy. Running Backs: There are plenty of RB's to be able to maintain this average, especially if you got Terrell Davis early. Heck, you don't even need 400 points from RB#2 and #3. No problem, almost halfway there. Kicker and Tight End: The 5th best each week is 93 and 118, but I'm not greedy, just give me ones that average 70 and 90, and there a five each of those out there. Small World must have miscalculated my score. OH NO, Wide Receivers. This looks hard. There are no WR's that can even match the 5th and 10th WR for the week. There are only four that can beat 15th for the week. Unless I made a whole lot of extra points in the previous positions, I need 687 points a week. Well, I've had A Freeman (9), J Rice (7), R Moss (5), E McCaffrey (4), T Glenn (4), Billy Davis (4), T Owens (3), T Brown (2) and R Brooks (1). Hope you didn't have them the weeks I did, because they've averaged 447 points a week for me (240 short times 13 = 3,120). In big trouble now. Team Defense: You wouldn't think picking one of the five best defenses of the week would be that hard. Forget the best, I'll settle for fifth. If you did, your defense would have averaged -216. Well, it's been tough for me. I guess at -359, I'm closer to 15th best for the week. (Let's see, 143 times 13 = 1,859). Guess my score is right, just way behind. Oh, for a few good defenses and wide receivers. That seems to me is where it's really going to make or break you. - Steve H. (12/4) response from Guru: What this probably really demonstrates is that the week-to-week volatility for WRs and Defenses is much greater than for other positions. I suppose that's consistent with intuition. So, since the "frontrunner standards" for those positions needs to be relaxed, I guess the other positions need to be tightened up. Here's another way to look at it. If you had the best YTD producers at each position (Young, Cunningham, Davis, Faulk, Anderson, Freeman, Moss, Owens, Sharpe, Anderson and Miami) and just held them throughout, this roster would have produced only about 25,240 through the first 13 weeks. Now, each of those players has had a bye which I didn't compensate for, but this team was also impossible to own throughout, since it would have cost $57 million on opening day. (Surprising how cheap it would have been isn't it?) - Guru Guru: I'd still be in the top 300 if it wasn't for the Arizona defense NOT PLAYING DEFENSE for the few weeks I had them. I also think that smallworld should have stuck with at least one wild-card player. It appears they based their prices on last year's model when we had two wild-card players and nobody had any cash left over. Ah, the good old days! - Jeff H. (12/5) response from Guru:I don't mind the high impact of the defense formula. In fact, it seems like analysis of defenses and matchups is one area where the proper amount of research and understanding of the game can make a difference. What I find troubling is that the pricing of defenses never really seems to be reflective of performance. Part of the reason is that initial prices and the initial draft have a significant impact on how high or low prices can ultimately go. So when Oakland started out as the cheapest team, it was fairly heavily owned on day one, and had limited subsequent upside, in spite of how well the unit has done this year. Also, given the importance of the defensive scoring, matchups and bye weeks have produced much more week-to-week shuffling of this roster slot, compared with individual player slots. This "churning" makes the price change mechanism ineffective as a means to bring prices to a more appropriate level. - Guru Guru: I like it the way they've done it. Including individual players would dramatically expand the list of potential players, and team defenses are much easier to evaluate. Plus, it's unlikely that, without distorting the scoring too much, individual defensive players would have much of an impact - sort of like kickers. - Guru Guru: I had all five of those guys as well, but fortunately, they were evenly split among my two teams, so the damage was diffused. I've thought about the issue you raised. Another way to look at it would be to see how a player performs following a big price change (in either direction). That's probably a similar question, since big price moves tend to follow extreme games - but it might be easier to measure, after adjusting for injuries (which I'd have to do in either case). In baseball, and even basketball to a lesser extent, many players tended to have hot streaks, and the trick was to jump in before the streak ended. But it's not at all clear that football is conducive to "hot streaks". I'll try to see if I can find some quantitative evidence. - Guru Guru: Smith started out so well last Thursday, too. He got all of his 76 SWP in the first quarter. - Guru Guru: Now with the limitless cash, I'm sure lots of teams carry Freeman, Moss, & Owens, but with Freeman's injury I expected an increase from Anthony this week. But instead, his price drops $90,000. Now he's only the 22nd scoring WR on the year, and there were some other WR's with big weeks this week (McDuffie, D Alexander, H Moore), but going against the soft Green Bay defense on Monday night, I think Anthony is a decent play. - Kevin L. (12/3) response from Guru:No question that Anthony is a bargain, but bargains are plentiful. For example, both Eric Moulds and Terry Glenn look cheaper. And as you note, franchise value inflation has become so rampant now that finding cheap players doesn't matter for a lot of teams. The quest is merely to find the best players from week to week, regardless of price. If you think Anthony is going to be one of the top three WRs this week, then go for him, and if you're right, you'll gain on the field in points, which is really the only contest that counts. But don't expect to harvest capital gains from heavily discounted players, because that just isn't likely unless the player has just come off a barnburner week - and that's usually a bit late to be picking him up. - Guru Guru: Home team scoring 22.6, visiting team 19.4 through week 13. 3.2 pt difference times 20 = 64 A couple weeks ago it was 3.4 and I've stayed with the +/-35, close enough. League wide this has stayed real close all year. The individual teams range from -16.9 (Oakland - not suprising) to -22.9 SWP's per points allowed per game. Also, I think teams that play in domes get used to the controlled climate, no wind, no mud, etc. I can't compare outside stadiums too much (the Eagles in Vet Stadium a couple of times back in the mid 70's are the only pro games I've seen outside, and the last game I was there, all I heard were boo's), but games I've been to at the Pontiac Silverdome and the Louisiana Superdome, when the fans get into it, you'd think the roof is going to come off. This has to have an effect. I'm sure outside stadiums, if the fans are really into it, has to mentally pump up the home team. - Steve H. (12/2) response from Guru: Guru: I dunno, Sam. The only way I get any compensation for this site is by "exposing" the advertising banners on my web pages. While it's not a particularly lucrative arrangement, it is at least "cash flow positive". If I were to send out a regular email with blurbs, articles, etc., it would circumvent the only revenue source I have. If it would be worth it to enough Gurupies, though, I might consider offering a service whereby I would send out a daily email with the blurb, and any current article and/or feedback items. I would charge for it, probably a modest amount, like $5 per sport, or maybe $10 for all sports for a year. It's not clear to me why this would be a benefit to anyone, since if you're getting email, it should be simple enough to scoot over to RotoGuru to see what's new. But if there is interest in this type of arrangement, I'd consider it. Any interested Gurupie should let me know if this sounds appealing; otherwise I'll just drop the idea. - Guru Guru: This is probably "too hard", but awarding the league average for the week to the bye team would also add a little intrigue. Would not be sure what you were getting. Couldn't be the worst, or the best. Just in the middle of the gamblers. - Steve H. (12/2) response from Guru:I agree that "forcing" a non-bye defense is a non-starter, administratively. Adjusting the point award is the way to go. I considered the idea of applying the actual average for the week, rather than a fixed value which was representative of the long term average. That could work, although I'm not sure it provides any more disincentive for holding a bye team than a flat charge of -420 SWP would do. If you want to provide a stronger disincentive, why not make it the average of the 15 worst teams of the week, or something like that? That would certainly do it! - Guru Guru: Never heard of it. Any of you Gurupies (among the ton out there) have anything to contribute? - Guru Guru: Actually, had SW gotten the math right, the bye week charge would have been more like -415 instead of -325. (Last year's average was really -415, and through 13 weeks, this year's average is -420.) The attractiveness of a bye week defense was that it was likely that -325 SWP would be a better than average performance. In fact, a look at the average defensive points for the teams in the RotoGuru rankings shows that the median average defensive score for these teams is around -350 per week. But less than a third of these teams have averaged better than -325. So while a -325 charge is certainly an attractive "safe haven", a bye week charge of more than -400 SWP would be a strong disincentive to go that route. - Guru Guru: Interesting twist, which I like. Yesterday, on one team I had the Jets (vs. Carolina), which went pretty well until late in the game, when the Jets had things well in hand. At that point, Carolina ground out 2 meaningless TDs, and the Jets were presumably in "prevent" mode, willing to allow clock-burning drives. So, it didn't turn out as well as expected. Some bonus points for a win would have made it more palatable. - Guru Click here for prior feedback letters, by month: November . . . . . October . . . . . September . . . . . August . . . . . July . . . . . June . . . . . May . . . . . April . . . . . March . . . . . RotoGuru is produced by Dave Hall (a.k.a. the Guru), an avid fantasy sports player. He is neither employed by nor compensated by any of the fantasy sports games discussed within this site, and all opinions expressed are solely his own. Questions or comments are welcome, and should be emailed to Guru<davehall@home.com>. |