August 20, 2018 at 9:09 pm
Comments posted to this topic are about the item Super Nerds
August 21, 2018 at 12:29 am
I stopped watching Formula one motor racing a long time ago. The microanalysis of performance had resulted in cars that made overtaking almost impossible and only a mistake in the pitlane could overturn the result of the dash to the 1st corner. Even a pitstop to change 4 tyres and refuel was down to 3 seconds!
Technology had negated the whole purpose of the sport. To thrill, provide entertainment and downtime for the audience.
Cycling was summed up for me by a journalist who said he didn't follow the sport to watch chemists compete.
How do we get the conflicting drivers of competition, funding and technology right? I don't know but sport that is boring defeats the object. Sport that requires huge outlay and dangerous obssession will alienate people from what should be one of life's pleasures
August 21, 2018 at 1:20 am
First of all, thanks Steve for the recommendation of the podcast 'Revisionist History'! It is my podcast of choice now on the train to/from work (when I'm not cycling).
For me, sports are something one plays or possibly supports, when my friends, children or school are playing. Congratulations, Steve, on giving up your time to coach. It is very decent of you.
When analysis at this level comes into a sport, it is no longer about enjoyment, rather it is about winning. I have long given up on the Olympics, Tour de France, professional football or anything else where you win or you die [1]. Anything that involves big money is little more that a statistics-generating machine in my books.
August 21, 2018 at 1:37 am
As many that follow your articles and in general following SQL Central, we are all data "fanatics" you could say. Data can be indeed helpful to make decisions of all kind, we all know that. IA, data analysis and everything surrounding information is becoming more and more automated, algorithms are used to predict more and more accurately the possible outcome of this or that situation. Algorithms already (in silence and behind curtains) mandate many decision in our daily lives, even for those that say "I make my own decisions". I would prefer a less analytical sport, more heart and feels. You said, there are less bunting, stealing and similar actions that precisely add to the emotion of the game. Same for road cycling as others mentioned, everything is about the numbers, Watts generated, potentiometer, hear rate and alike, less and less riders try to do their own race, based on their feelings, is not longer fun to watch. That's why I still follow mountainbiking and I'm myself an amateur racer, technology helps, but when gravity is on play, on the downhill sections, there is only you, your bike and your skills.
August 21, 2018 at 7:15 am
It seems an open question as to whether this can really apply to American football. There is a view that you could spend a lot less on players in the lower drafts that will then work as a team. But a top team needs to have a spark. Not that it has no uses. Only that it doesn't seem to apply to the same extent as baseball.
August 21, 2018 at 7:41 am
Ultimately it is the managers role in baseball and the coach in football to make the decision as to which players are going to give you the best chance to win.
Stats are averages... they do not guarantee any outcome. If player bats .375 against a right-hander and .250 against a left-hander, it doesn't mean that they won't get a hit (or maybe even a home run) against a left-hander, it just means that they are less likely to do so "on average".
When I was fresh out of college I tried to write a program that would predict NFL scores based on who was playing, home field advantage, field conditions, etc. One thing that I never could account for is the instance when a player plays way over their head. A running back that has a breakout game, or a safety that picks off 3 passes in a game after not getting any interceptions all season. These outliers become a part of the stats base only after they happen.
I heard that College football was doing a study to see which players had better games when their families were in the stands, and which ones got nervous and played better when their families were not present. Patterns like this could make a huge difference when losing a single game could cost you a shot at the national championship.
If the players are complaining, it's because they do not have the trust of the manager that they will produce in a particular situation. I have seen many times where the statistical choice would be one direction, and yet the manager goes with their guts and it pays off. In hockey you go with the hot goalie until they cool off.
So for the players to complain... easiest way to fix that is to raise your stats so that you are statistically the guy to go to in any situation. Either that or play well enough to give the manager/coach that gut feeling that statistically you may not be the best option, you will produce when asked to.
August 21, 2018 at 10:19 am
It's definitely a thorny issue to use stats in sports. The thing many people want in sports is some guarantee. This doesn't do that. You can't predict a score, but across a bunch of games, you might be better at picking the winner. You might give yourself a better chance, which isn't the same as a win.
If you look at this chart of Kevin Durant's made shots, what do you see? http://assets.sbnation.com/assets/2129919/durant-chart2.jpg
You obviously want to force him to the right side. That doesn't mean if everyone does that he will lose or miss. The next set of shots might just reverse the chart. Athletes adapt, especially great ones.
In baseball, skipping the bunt for a hit isn't going to necessarily win you this game. Batter might fly out, hit into a dp, etc. What the stats show is that over time, at this level, it's more likely that hits will outperform the loss of the out. Across a season, not a series or game. In any short period anything can happen.
In football, there are tendencies for players to go one way or the other. Some players don't adapt as well. QBs don't look at all receivers the same. The blockers use certain techniques. You learn those things, not to beat them every time, but to be ready at times to change how to approach a play.
Same things in all sports, but you have to look long term.
In terms of picking players, some of this matters, but it's harder because the same at the pro level can be different than at lower levels, especially for American football.
August 21, 2018 at 10:21 am
Thanks, it's at the competitive level, so there's a little compensation, but that's mostly because my daughter still plays. I've set up my team to play opposite hers, so from Jan-Apr, I'm likely buried every weekend with volleyball.
Next year I might do a younger age/league as I like working with the younger kids rather than the older ones.
August 22, 2018 at 7:01 am
I wonder if the use of statistics like the editorial talks about is inevitable in any professional-level sport. After all, bonuses and jobs depend on winning, so the team (read more as management) will take any edge they can get to win. As time goes on, newer, and better, and more detailed stats can be collected and analyzed, potentially to the point where a team will be able to say "Joe hits a .370 against right-handed pitchers except when the humidity is over 30% and the air temp is 83F with falling mercury, so lets put in the pinch-hitter here."
Power sports (F1, Indycar, NASCAR) I think it's less about the micro-analyzed performance stats, as it is the governing bodies dictating the specs of the vehicles themselves. It's gone to such an extent that NASCAR teams keep finding new ways to "cheat" around the rules (bit of clear tape on the spoiler for a bit more downforce, a "broken" support in the rear window to let it bow inward giving smoother airflow over the spoiler, hip-checking the quarter panel to dent it and alter airflow, the list goes on.) Largely, I think most professional motorsports are turning more into "spec racing" than "technological innovation racing."
Arguably in all cases, I think the lower-levels of the various sports are probably more fun to watch, simply because it's less analyzed (at the team level.) So, more chances are taken, and frankly, mistakes made (giving chances to the other team.)
And a reply for Dave Poole: An article I read a couple weeks back on a car site was also lamenting the "elephants on parade" nature of modern F1 racing. The suggestion the author had was to make the cars *smaller* in size, as the current cars are so large that on some of the classic courses (Monaco was explicitly mentioned) there's virtually nowhere for one car to pass another without hitting the wall...
August 22, 2018 at 7:43 am
Bunting and stealing are highly effective, if the individual player is a string runner.. and especially if the other team doesn't see it coming (ie: they read the memo that it's not effective and assumed the other team wouldn't try it). Some of the best investors on Wall Street make their fortune by engaging in strategies that run opposite to the general advice offered in the mainstream financial news media.
"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho
August 22, 2018 at 12:46 pm
Eric M Russell - Wednesday, August 22, 2018 7:43 AMBunting and stealing are highly effective, if the individual player is a string runner.. and especially if the other team doesn't see it coming (ie: they read the memo that it's not effective and assumed the other team wouldn't try it). Some of the best investors on Wall Street make their fortune by engaging in strategies that run opposite to the general advice offered in the mainstream financial news media.
Not in a percentage case. Even looking back at greats like Ricky Henderson, you can find that the team might have been better if he had let the hitters hit behind him. In rare cases it stands out, but baseball is a game of percentages overall.
I think they'll be more effective again once they drop to some level where they rarely occur. For now, the geeks are super good at catching tells and places to run, and catchers are crazy quick these days, so stealing and bunting don't work as well. It's also less about if the batter gets to first. The trade of the out isn't usually a good call.
That being said, pitchers are doing great at striking players out this year, so maybe we'll see more bunts come back
August 22, 2018 at 12:57 pm
I remember reading an article that discussed handicapping horses. They said that they originally gave the handicappers a ton of information and the results were all over the board. Then they reduced the set of stats to about 10 metrics, and all of the handicappers did much better.
I believe that the point of the article was that you needed the right information to make a good decision. Sometimes there is more information than you need and it clouds the big picture.
Then again, maybe the model that they were using just did not account properly for the additional factors?
Viewing 12 posts - 1 through 11 (of 11 total)
You must be logged in to reply to this topic. Login to reply
