Tag Archives: prediction theory

Debunking the Momentum Myth

This post first went up back in March 2010. I’m re-posting it during this off week because it’s assertions remain valid. Hopefully it’ll stir up some discussion! —Jed Henson
______________________________________________________

“Kevin Harvick has run well for the past three weeks, and that gives him momentum going into this weekend. Put him on your roster.”

I see recommendations like this in lots of fantasy NASCAR blogs, forums, and articles written by experts. Dan Beaver, a columnist for Yahoo! Sports Fantasy Auto Racing, seems especially smitten with momentum. Pretty much every race preview he writes for Yahoo!’s fantasy players contains recommendations based on driver momentum.

So, what’s this all about? Is there some sort of mysterious force at work that prolongs lucky streaks? Do drivers on a hot streak gain some sort of mental advantage over their competitors? Does success on one type of track beget success on another?

The Luck Factor
Let’s start with a simple game that does not have any human input. Say you flip a coin and it comes up heads five times in a row. On the sixth flip, does momentum make it more likely to come up heads again? Or less likely? Neither, of course. The odds are 50-50 every single time you flip the coin no matter what happened on the previous flips. In a pure, luck-based game with zero human input, momentum does not exist. Continue reading

Yahoo! Fantasy Tips & a Yahoo! Fantasy Tool

Over in the archive at FantasyNascarPreview.com, there’s a pretty good strategy article for Yahoo! Fantasy players. Titled “Yahoo! Fantasy NASCAR Strategy,” it first explains how the Yahoo! game works, then goes into some tips and strategy. I agree with most of the author’s points, and his section on Group B strategy is really good.

I do have a couple of quibbles, though. First, while I agree that we should look for situations in which we can save a big gun’s starts for his top tracks (e.g., Jimmie Johnson on 1.5 milers), I think we should temper our start-saving aggressiveness early in a season. At this point in the 2010 season, for example, the series has run on only one short-flat track (Martinsville), and it’s not quite clear yet that the big guns and duds from last year on short-flat tracks will also be big guns and duds on those tracks this year. I’d hate to aggressively save 2010 short-flat starts for a 2009 short-flat master only to discover later this year he’s no longer a short-flat master. Continue reading

Feedback on the Sprint Cup’s New Spoiler

The 18 heads out for a test run today at Charlotte.

The 18 heads out for a run during testing yesterday at Charlotte Motor Speedway.

Sprint Cup teams wrapped up their testing of the new spoiler this afternoon at Charlotte Motor Speedway, and while just how significant an impact it will have remains unknown, the general consensus seems to be that it won’t be a big impact. I’ve seen several informative articles on the topic, and I’ll list some of the better ones below:

Johnson Not Worried About Spoiler,” by Jeff Owens at SceneDaily.com. Go to SceneDaily.com for a bunch more good stuff on the spoiler.

Gordon: Spoiler Not a Big Change,” by Mike Hembree at SpeedTV.com;

Drivers Take Spoiler for a Spin at Charlotte,” by David Newton at ESPN.com;

Spoiler Impact Unpredictable,” by Mike Hembree at SpeedTV.com;

Note: I found most of these articles via Nascapper.com’s NASCAR News forum.

The Sprint Cup Tire & Race Predictability

In my never-ending quest to get better at identifying which Sprint Cup drivers will run well each week, I’ve been reading up on the tires Goodyear provides the race teams and trying to figure out if and how I should add a tire factor into my weekly analysis. I’m kinda stumped, and unfortunately I don’t have access to someone like Chad Knaus to turn to for enlightenment. So, I figure I’ll turn to you, the readers, and see if we can get a discussion going. To that end, I’ve posted two questions below.

Question 1
It seems clear changes in tire construction and compounds affect a NASCAR Sprint Cup car’s handling. Just changing the air pressure in a tire changes the handling–that’s actually a primary set-up tuning tool. The question is, how much does a new/different-type tire affect the handling?

Let’s say you take the exact car Kyle Busch won with at last year’s Bristol race on a clear, 70-degree F day and run 50 laps at Bristol with a fresh set of the tires he won during that race. Then, without changing a thing (including the weather), you bolt on a fresh set of the tires the series will run this weekend (they’re different, correct?) and run another 50 laps. How much differently will the car run? Slower through all 50 laps? Faster through all 50? Maybe the same at first, but then fall off quicker?

I imagine there are a zillion variables that play into this, but if we can get at least some sense of how significant a tire change is, we’ll gain a better understanding of how much a tire change decreases a race’s predictability.

Question 2
How can the fan, blogger, fantasy player, etc., figure out if Goodyear is indeed bringing a tire to a race that differs from the tire used in the previous race at that track? I get the impression it happens just about every race, but that’s based soley on what I hear from Darrell Waltrip, Larry McReynolds, etc., during the practice telecasts. Are there any reliable, consistent NASCAR tire information sources out there?

2 Recent Tire Articles
This week I came across two tire-related articles I found pretty interesting. The first, “Goodyear Gets a Grip in Cup Series” by David Newton at ESPN.com, discusses the strides Goodyear has made since the tire debacle at the 2008 NASCAR Sprint Cup race in Indianapolis. The second, “Goodyear Finally Has a Grip… Or Do They?” by Bryan Davis Keith at Frontstretch.com, provides some counterpoint to Newton’s piece, and raises the alarm that perhaps Goodyear has not learned its lesson.

I’d love your feedback, thoughts, etc., on the questions and/or the articles.



NASCAR Track Guide

I constantly work on improving my ability to predict NASCAR Sprint Cup race results, and when I read an article by Christopher Harris on ESPN.com a while back, I made a big leap forward in my understanding of this sport. Titled “NASCAR: Track and Driver Guide,” the article discusses how the tracks on the circuit vary and what the fantasy league player should take into consideration when trying to identify who will run well at a particular race. Harris even includes a table that lists the length and banking of each track.

You can get to the article by clicking here.