SB Nation User Blog: pizzacutter

Ten statisticians you should know about

pizzacutter — Thu, 12 Nov 2009 03:23:43 -0000

Ten statisticians you should know about

All of these cool statistical techniques didn't just magically appear, y'know.

The "30 parks on a budget" challenge

pizzacutter — Mon, 12 Oct 2009 00:50:45 -0000

The World Series is over. You're looking at another off-season of trying to figure out if your team should sign some middle-of-the-road middle infielder who won't bring the pennant next year either. And because you can only do that for about 15 minutes before crying, you need something to do that can keep your mind on baseball, so that you don't have to resort to *gasp!* watching the NFL, NBA, NHL, EPL (go Everton!), WWE, or SCOTUS. (Tonight on Monday Night Litigation, it's Sotomayor vs. Alito. Justice will prevail!)

No, let's do something baseball-related. Let's plan a baseball road-trip, the best baseball road trip ever. And let's do it on a budget.

Your mission is simple: In 2010, you will (in your imagination) travel to all 30 MLB parks, see 30 games, and you will figure out how you would do it at the lowest possible cost. At least in theory. (It's a free country, so if you actually want to do this in real life, you can... but the point is that it's a fun mental exercise.) Using the real 2010 MLB schedule, and some other web-based resources, you will put together an itinerary that would allow you to meet your goal. The person who can do it in the lowest amount of cash will win a cookie and bragging rights. (Note: no actual cookie will be awarded.)

I will tell you, this "game" is horribly addictive. Worse than Farmville. You won't think so at first, but it is.

Intrigued? The rules are below...

The goal: Using the real 2010 MLB schedule, you will create a tour that will take you to a regular season game at all 30 MLB parks. Obviously, you'll see all 30 teams at least once. The visiting teams are unimportant and if you end up seeing the same visiting team three or four times during the season... well, you'll just get to know that team a lot better. You just have to follow...

A few general rules:

You are a total loser and have no friends anywhere in the USA. It's tempting to say "Well, when I get to Chicago, I'll just bunk out with my old college roommate who lives near Wrigley and walk down to the game. And when I get to Phoenix for the D'Backs game, I'll just barge in on my aunt who lives out there. Even though I never talk to my aunt." That's not the point. And you should send your aunt a holiday card.
You now own a car. Congratulations if this is your first car. It's a red one. You may or may not end up using it.
You get one freebie city, and it's the MLB city which you are sitting closest to right now. If you live in New York, Chicago, San Francisco, or Los Angeles/Anaheim/California/whatever the Angels are calling themselves these days, read on. In that city, you do not have to stay in a hotel or pay for food. You can stay in mom's basement and eat her home cooking. In other words, what you're doing already.

Getting started: You may start your tour at any point in the MLB schedule, but you have to be able to say, "I'll start at the Team A/Team B game on this date." You can take as long as you want, although, as you'll see in a moment, it behooves you to be faster. However, you must either start and end your trip in the MLB city that is geographically closest to you. This is for those who live in eastern Montana. You can magically start the trip in Minneapolis without having to factor in the trip from Montana to Minnesota.

If your closest MLB city is a two-team city (see the four above), your first game must be for one of the teams in your city (so either Mets/Yankees, Cubs/White Sox, Angels/Dodgers, or A's/Giants... and let's not pretend that Oakland is a separate city) and the last game must be the other of those teams. Your pick as to which is which, but for the last game, you have to treat it like you are on the road and book a hotel, etc. This is because in the course of the month to month and a half that you were away, your friends and family members realized how much they hated you all along and will not help you out at all when you get back.

Getting there, part one (city to city): The biggest issue in a trip like this (and maybe the whole point) is getting from city to city. You have a few options.

You can drive your own car around the country for all or part of the trip. If you do that, you have to live by a few rules. First off, your car has to find its way back home, so if you're starting in Atlanta and you use the car to go to Tampa for your stop there but then catch a flight to the next part of your trip, you eventually need to come back to Tampa and drive the car back to Atlanta. However, for each day that you leave your car in your non-home city, it's $10 for parking/car storage. Also, your car can not magically follow you, and since you have no friends, you can't just have Larry drive the car to the next city to meet you there. Even if you did have a friend named Larry, he wouldn't like you that much.
Your car gets 30 mpg (city or highway) and gas costs $2.40 per gallon. Yes, I know gas prices vary widely across the country. But we need a standard to make things fair. So, each mile costs you eight cents to drive. All distances will be measured by the Mapquest door-to-door road distance. If you want to turn it to "shortest distance," that's fine. Then again, "quickest route" might help you in other situations. We'll pretend that you won't get lost. Somewhere out there, my wife just snickered.
You could rent a car. You have to find a published rate at one of the major rental companies in the city you'll be in and tack that on to your running toll, plus pay the 8 cents per mile in gas. If it's a one-way rental (you'll rent the car in Boston and drop it off in Philly, you have to get a quote based on that.)
You could take the bus/train. If you can find a company that runs a regular route between those cities (no personal charters!), then you can take to the roads that way. However, you have to abide by the schedule that they set. So, if their bus/train pulls into the station at 4:00 am... think about that for a minute... do you really want to be at a bus station at 4:00 am. You have to be at the bus station 30 minutes before the bus leaves.
You could fly. Again, you have to find a published rate on a flight, but you have to abide by their schedule. If you are flying, you have to be to the airport an hour and a half before the flight is scheduled to depart. The flight can be on any airline. And you have to use the full price of the ticket after adding in fees and charges. They all have to list that now.
Some folks might want to fly to a "secondary airport" because sometimes the prices are cheaper. (For example, flying into Canton/Akron "airport" and then driving to Cleveland.) That's fine, but factor in the rental car/bus to Cleveland. Hey, it might save you a few bucks and this is a competition of doing this in the least amount of cash.
As much as I love William Shatner, no, you may not use Priceline and say "well, I'll just get a flight for $25." You have to go for the pre-published rates. You can, however, use any of the major travel sites to help you do your research and fly whatever airline will get you there the cheapest. And no, you can't use frequent flyer miles. This is also cheating.

Getting there, part two (within the city): Once you are in the city, you have a few options for getting around. You will need to book a hotel, as you have no friends. You may select any hotel you like anywhere you like, again so long as the rate is published. Obviously, this will go toward the final tab for the trip.

If you're driving into town in your own car (or renting a car, most likely at the airport), you can drive right to the hotel door and park in their parking lot. However, some hotels have parking fees. Choose carefully. Hostels are fine... if you like waking up in a room full of European tourists who speak about as much English as you speak of Portuguese. Depending on the age and gender composition of that group and your own preferences on the subject... well... there might be kids reading this.
if your hotel has a free shuttle from the airport to their front door, bully for you!
No shuttle? At the airport, you can usually get one of those "hotel vans" to take you where you're going. Nothing like a 12-passenger van that smells like a diesel leak and has you crammed in nice and tight. Those come with a cost though too, but you can shop around on those.
Taxis are $2.00 per mile. Again, the Mapquest door-to-door distance rules. Never insult the barber before the haircut, but stiffing the cabbie the tip after the ride is optional.
Walking is free, but you can only walk 4 miles total per day, and you walk at 4 mph.
Now, not only do you have to get to the hotel, you also have to get to the ballpark and possibly back. Public transportation is allowed. You need to factor in the cost to ride it, but if you can show a route that goes from your hotel to the ballpark and back (you can also use it from the airport/bus station), that's fine. Again, you are at the mercy of their time-table. Also, if you need to walk to the stop/station, you have to deduct that from your four mile walking allowance.

The reason that you're doing this, the games: You must attend 30 games. And that means that you'll need a ticket to get in. You're welcome to pick the cheap seats (this is after all about doing things on the cheap), but really, if you ever actually do this, it means you have way too much money to burn. Spring for some good seats!

Remember that some teams price different types of games differently. For example, some teams are cheaper on the weeknights in May than the Saturday in July.

You must list the 30 games that you'll be attending, and they must be actual games that are scheduled to be played. Don't worry about rain (or if you're touring the North in April, snow). Again, it's your choice as to when you would start. However, you can't go home. Once you start, you have to complete all 30 games before setting foot back at your front door. So none of this "I'd hit Dallas, Houston, Phoenix, San Diego, and Colorado" then fly home to Kansas City for a few days and eat free food, and then head out again..." What you're really doing is buying yourself an extra day of free accommodations because you know you'll need an off-day on the road. That's cheating. You can stop through along the way, but no lingering.

Other costs: For every day that you are on the road, add $25 in per diem/food expenses. This applies even if you don't go to see any games. You still gotta eat. (And yes, someone out there is saying, "But I can live on five dollar footlongs." I say to you man does not live on bread and cold cuts alone, but on every word that comes from the mouth of Joe Morg..." I think I'd better stop there before I get smote.

I'll assume that you have no baggage costs on planes. Ever since airlines started charging per checked bag, everyone pulls their roll-aboards onto the plane and gate checks them, which slows down boarding by about ten minutes per flight. Not that I'm bitter. I'll assume you're living out of your carry on. And that you brought at least 30 pairs of clean underwear. Or that you found a random laundromat on an off-day.

The time element: This could get fun. Not only are you competing against your wallet, you have to actually make this a believable itinerary. So, if your flight lands at 6:55 in Pittsburgh, and first pitch is at 7:05, you're not going to make that game, so don't try to pretend that you will. You are not magic. There are several time restrictions in this exercise by which you must abide.

As previously mentioned, any time you are taking mass transit of some sort, you must abide by their published time table. Most public transportation systems have on their website the ability to get door to door times for any trip. Your walking speed is 4 mph.
If you are in a car of some sort, the Mapquest "time to travel" will rule all questions of how long it will take to get there. This includes taxis.
For any airplane flight, you must be able to make it to the airport an hour and a half before the slated departure time. In happy news, we'll pretend that none of your flights will be delayed and will all land exactly when they say they will. Isn't that a hlarious sentence? For a bus or train, you need a half hour of cushion.
You will also be "stuck" at the airport for 15 minutes after a flight lands if you are taking public transportation or hailing a taxi, and 30 minutes for car rentals and hotel shuttles (free or paid.)
You can't check into a hotel before 3:00, but you should do so before going to the game. An itinerary item that says "2:38-3:00 Screw around in area near hotel" is fine if you just need to kill time. For a 7:00 night game, that should be fine. But for a 1:00 day game, this could be more interesting, because you must also check out by noon. If you're catching a flight after the game, you can check your bag at the hotel for $5.00, but you have to go back to retrieve it and pay the costs associated with getting back to the hotel.
You must be at the ballpark 30 minutes before the start of the game. Why? You're doing a tour of ballparks! You want to be able to spend some time walking around the park and taking the obligatory "This is me next to the sign that says Welcome to ~~Enron Field~~ Minute Maid Park" picture, right? Earlier is fine.
American League games last 4 hours from first pitch time (this includes the time to get out of the stadium, back to the parking lot/train station/bus stop). National League games are 3.5 hours long. Interleague games go by home park. Some of them might go extra innings, but we can't plan for that now.
You must get 8 uninterrupted hours of sleep every day. No sleeping on the plane. Or on the bus. (Trust me on the bus.) Only at the hotel. You can pick your hours either by preference or around your travel plans. You must also allow one hour after wake up time before you can check out of the hotel. You need a shower. (Trust me on the shower.)

That's a lot of rules. If there are disputes, post them in the comments. I'll be happy to rule on them/clarify them.

So what should this look like? Let me give you a sample of what I have in mind.

April 5th, 2010

Time	Event	Distance	Cost	Link
	Per Diem		$25.00
7:11 - 8:08 am	Belmont and Sheffield to MDW (red and orange lines)		$2.25	link
8:15 am	Arrive MDW
9:45 am - 2:29 pm	MDW to HOU, Delta 1567/879 (connect in ATL)		$118.20	link
2:44 pm - 3:04 pm	Hobby Airport to Howard Johnson Express Inn via taxi	14.07 mi	$28.14	link
3:04 pm	Check into HJ Express		$48.37
6:01 - 6:25 pm	Houston Metro Routes 36, 121 to Minute Maid Field	0.61 mi walk	$1.25	link
6:35 - 10:35 pm	Giants at Astros, 7:05 pm first pitch	--	$7.00	link
10:35 pm - 10:59 pm	Houston Metro Routes 121, 36 to HJ Express	0.61 mi walk	$1.25
11:00 pm - 7:00 pm	Sleep

Belmont and Sheffield is near where I used to live in Chicago so I'm using that as my starting place. Let's say I'm starting out from there and going to Houston for my first game of the trip. (Why Houston? Why not?) I've accounted for my transportation and lodging and put together a realistic time frame, plus accounted for all my expenses.

I didn't do a thorough job researching this. I could probably find a cheaper way to get to the hotel than taxi, and maybe there's a bus route that would have gotten me there cheaper than the flight. I also should really have put up the page that had the price quote for the hotel, but I closed that window.

Now, I'm in Houston with no car and 29 more parks to see. But I did get to see Opening Day in Minute Maid Field. Probably Tim Lincecum vs. Roy Oswalt. And in doing so, I spent a couple hundred bucks. Could I have done it cheaper? Probably. Can you?

If you do accept the challenge, post it here as a FanPost or link it off your own blog or something. I'd love to see if anyone can actually put one of these together, and if there are several, who can do it cheapest.

Which tells us more: The last 7 at bats or 7 at bats against this pitcher?

pizzacutter — Sun, 18 Oct 2009 22:46:18 -0000

Suppose that a batter is 5-for-7 over the course of his last seven plate appearances. Pretty good two-game stretch, eh? Now, suppose that another player has a career 5-for-7 mark against the day's starter, going back a few years. (You could substitute "the reliever who was just brought in" if you like.) Which is the more meaningful, the last seven PA's chronologically, or those seven PA's against the pitcher?

The proper answer to this is "Who cares? Seven plate appearances isn't enough of a sample to tell us much of anything." Let's leave that aside for a moment. Let's enter the magical world inhabited by magical gnomes, Joe Morgan, and more MLB managers than I care to mention. Let's pretend that you can actually draw serious inferences about future performance from seven plate app... I almost got through that with a straight face.

The bigger question is which is the better source of information: performance in the recent past against a different set of pitchers or performance against this specific pitcher over an array of years, given an equal number of plate apperances in each case. I suppose it's an empirical question and one that I haven't really seen a study of directly. I've seen a few things that lead me to believe that neither is a particularly good predictor, and so the "winner" between those two would be the winner of the "tallest grasshopper" award.

I don't care so much about which really is the better predictor right now. What amazes me is that managers, I presume in the absence of systematic study, have very clearly made up their minds. Consider how often you see managers make strategic personnel decisions (whether pinch hitting or moving someone around in a lineup or starting a guy who's normally on the bench) because "he has had success against Smith in the past." It's rare-er to see a manager do something like move a player up in the lineup because he's had a couple of good days in a row. (It does happen... just not as much.) Managers seem to value pitcher vs. hitter matchup stats. Now, if the reason that our batter had such great success against Smith is that Smith is a right-handed fastball guy and our batter eats those guys for breakfast, then that's fine. Of course, the problem is that over 7 PA's, anyone can get lucky against anyone, and to interpret that as anything meaningful is just silly.

But if it's just a matter of not understanding small sample sizes, we should see a great deal more re-shuffling of lineups based on recent performance. Managers seem to know, whether explicitly or intuitively, that the last seven PA's don't tell you too much about a player and so there's no need to change your strategy based on that. Why is it that managers understand small sample sizes in one situation, but not in another? There has to be another force at work.

I think there's something special in the manager's mind when it's batter vs. pitcher, and I think it's something that's a systematic error in thought across a lot of what passes for baseball analysis in the public sphere, something I like to call the virtus fallacy. Virtus is a Latin word which is hard to translate properly. It is the immediate predecessor of the English word "virtue", but in Latin, there's an added dimension. Consider that the Latin word for man is vir. Virtus was a mixture of courage, power, and honor that a Roman male citizen strove for. It was a characteristic that was "demonstrated" by some noteworthy act, often athletic or military. It might even be best translated as "manliness." The fact that you did well showed that you had virtus and if you lost, well... you just weren't the virtus kind of guy.

Following this logic, if a batter goes 5-for-7 against a pitcher, it must be because the batter is more of a man than the pitcher. And while it might be silly to believe that a plate appearance from five years ago when both pitcher and batter were five years younger in body and skill level would have a bearing on what's about to happen here, people generally assume that virtue is unchanging. If you were more virtuous five years ago, you're probably still the better man today. So, I should pinch hit you in this key situation. If a batter is 5-for-7 over his last few at bats, he probably did it against four different pitchers, so there's no chance for him to "prove" that he is more virtuous than any of those pitchers. He might be (and probably is) a good player or maybe he just had a good couple of days. But he's not proven that he's better than a specific other man.

So, the next time you see a manager base a key decision clearly on batter vs. pitcher matchup data (and make a hilarious strategic blunder), it might not be evidence that he is statistically challenged. Well, it is... just not in the way that you think. He's shown in the past by what he doesn't do that he doesn't make decisions based on micro sample sizes. It's just that his threshhold for accepting a "big enough to be relevant" sample size goes way down when the information is framed as a measure of who is somehow a "better man."

What would a perfect starter be worth?

pizzacutter — Wed, 16 Sep 2009 21:56:58 -0000

A small thought experiment:

Imagine for a moment the perfect starting pitcher. And when I say perfect, I mean he's practically perfect in every way. He is guaranteed to go out every fifth game and to throw nine perfect innings, striking out all 27 batters on 81 pitches. Guaranteed. If your team has a minimally competent offense (must... resist... Royals joke) and can squeeze out one run over nine innings, you have a win. We'll call him Pedro Halladay "Three Finger" Gibson-Young. Figure that over a 162 game season, he'll start 35 times (plus his inevitable All-Star Game start) and garner you 35 wins. If your team makes the playoffs, he'll do the same thing there in two games of each series. Need someone to pitch Game Seven? He'd make a rather nice option.

Now it so happens that this winter, Pedro Halladay "Three Finger" Gibson-Young will be a free agent this winter. Imagine that you have a $100M payroll to work with. How much would you offer him on a per annum basis? Remember that you do have 24 other roster spots to fill and every million that you offer to Mr. Gibson-Young is another million that you can't spend on the rest of the team.

Got a number in your head? (Really? That much? Since we're suspending disbelief, I gave myself the ability to read your mind. And yes, I heard that too. That's disgusting. I would never do that with a lobster.)

Two questions:

Are you paying a little extra for the Game Seven thing? Should you?
Consider the number that you just generated. Consider numbers that are thrown around for mere mortals who do not throw perfect games every time out and are not guaranteed to even be good. Are your numbers close?

Discuss.