8.29.2014

Luyman's Terms #14: Rabble Red



Today, I just want to show you guys my version of the "Rabble Red" deck that become so popular recently. I've seen quite a few different lists, and to be honest I'm not sure which list is the best. Mine is set up to still be legal post rotate, mostly because 99% of my cards are from Theros block and M15. Here is the list:
There are a few cards that I have in my list that I have not seen on any others, and I want to take some time and tell you why they are included:
  • Satyr Firedancer is a great way to burn out my opponents early drop creatures if they are trying to out-aggro me, and even a great way to get rid of big late game creatures. There have been a couple instances where I have burned my opponent for 5-7 with various cards and taken out a high toughness creature in the process. If one of those cards is a searing blood I can usually take out another creature as well. This almost always leads to a victory.
  • Frenzied Goblin is also nice to have in almost any stage of the game. It's a one drop creature that I can start turning sideways quickly and its ability is quite relevant in the late game when I'm trying to get in the last few points of damage and I don't have my Firedancer/burn spell combo.
  • Act on Impulse is a card good only in the late game, but it does its job very well. When I first started playing this deck I found myself without a hand after turn 5 or 6 fairly often. My opponent would be at 2 and I would be begging for the top of my deck to be a burn spell. Since I've put Act on Impulse in, I've not once had this problem. There have been a couple times where I draw the card in my opening hand or early in the game and it sits in my hand while I die, but I think that is acceptable given that the number of times it has drawn me that last burn spell I need is much bigger than the times it just sits in my hand. 
This deck has been a blast for me to play. My win percentage with it is over 70%! Barring some new amazing archetype out of Khans of Tarkir, I will probably be playing Rabble Red for the foreseeable future. What do you guys think? Could my list use a tweak here or there? Do you wish this deck would die? Let me know!

8.27.2014

Luyman's Terms #13: When Pinch Hitting Works

For this post, I will be using data from the 2008 and 2009 seasons because it is the most recent data I have at the moment. I am having issues updating my database and I hope to have more recent data in the near future.

When I first sat down to write this, all that I knew was that I wanted to do some sort of a post on pinch hitting. Pinch hitting strikes me as one of the most complicated parts of a baseball game, while also being one of simplest. If you're pitcher is having a bad day on the mound and it's his turn in the order to bat, you pinch hit for him. If the home team is down in the bottom of the ninth and it's the pitcher's turn to bat, you pinch hit for him. These are both easy decisions.

What complicates matters is if a team is down one or two runs late in the game, or even if a team is down five or six runs. What if you're down 6-1 in the 5th and the pitcher's spot is up? Is it worth it to pinch hit in such a low leverage situation? What if you're down 1-0 in the 7th and it's the pitchers turn to bat? Do you pinch hit for the pitcher to try to score and hand the ball off to a potentially worse pitcher? 

In all of the following charts, sheet one will be reserved for starters, sheet two will be reserved for pinch hitters, and sheet three will be your cheat sheet for determining which hitter type was worth more runs. According to my data, there were a total of 9718 plate appearances by pinch hitters in those two seasons opposed to 358137 plate appearances by players that started the game. This means that a little over 2.6% of plate appearances were by pinch hitters.

Here is the expected number of runs generated by hitters while hitting in a given out state:

 

It makes sense to see that the runs generated goes down as the number of outs goes up. If you have more outs, then you have fewer chances to score runs. What's interesting is that pinch hitters are worth more than non-pinch hitters in situations where they come to the plate with one or two outs recorded, but starters are worth more when no outs have been recorded. Let's see if we can figure out why.

To do this, we will have to break our data into more, smaller pieces. With smaller pieces it will be easier for us to identify exactly when/why our data shifts. Lets take a look at how  Here is the expected number of runs generated by hitters separated by base/out state:



According to this, there are four instances (out of a possible 24) where pinch hitters are worth more than non-pinch hitters when they come to the plate with no outs, but only two instances when they come to the plate with one or two outs. This seems backward since we determined that pinch hitters were worth more when they came to the plate with one or two outs in the first chart. These tables seem to tell us the opposite. Lets take a look at the data from a different perspective. What do the number look like when we take into account how many runs a team is down or if the score is tied when they send a pinch runner to the plate?





It should be noted that this is far from a complete chart. The score of a game of baseball often differs by more than four runs, but these situations were by far the most common, and I didn't want the results skewed by a small sample size. In every situation but three, pinch hitters were worth more than starters. Two of those three situations are when the batter comes to the plate with no outs, which helps confirm our first chart. It looks like its a good idea for a team to pinch hit if they are in a position to do so. What about when a players team is up when a that player pinch hits?




Same disclaimer as above, these were simply the most common situations and I didn't want the results skewed by a small sample size. Here, we see similar results to when a team is down when they pinch hit. The only situation in which a starter was worth more was when he came up to bat with no outs recorded, which also helps to confirm our first chart. It also seems as if even if a team is up it is advantageous for them to send a pinch hitter to the plate if they are in a position to do so.

So what about that second chart that seems to contradict everything else? Lets take another look at the data. This time, sheet three will show the number of times a pinch hitter came up to bat in those situations:



Now this is interesting. Most notably when a pinch hitter came up to the plate with runners on 1st and 2nd and one or two recorded outs. As you can see, there is a sizable increase in pinch hitters in those situations. Those also happen to be the situations in which pinch hitters were more valuable than starters. This may simply be a coincidence, but nonetheless explains why that chart seems to contradict all of the others.

According to all of this data, if a team has an opportunity to pinch hit, they probably should. It seems as if the fresh bat is worth more than the tired bat, especially as pressure increases due to outs being recorded or if the game is close. That said, I'm not convinced that this data is at all definitive. It might be worth looking into how much the handedness of a pinch hitter compares to the handedness of the batter he is replacing as well as how much pinch hitting for a pitcher vs pinch hitting for a position player changes the data. I think this is a good stopping point for today though and will give us something to chew on for awhile.

8.25.2014

Link City #2

Here's a list of things that I found interesting this week!

Baseball

Jeff Sullivan attempts to explain how the Orioles are beating their projections. Unsurprisingly, he fails. That said, the Orioles are having a heck of a year. And while I'm not a huge fan of results-oriented thinking (I do not think that being in first place makes the Orioles a good team, or vice versa), results are what matter and the Orioles have them.

Drew Fairservice takes a look at Mike Trout and why his value on the basepaths has taken a dip this season. It's interesting to look back on Trout's (short) carrer, and how he has changed so much as a hitter. He's no longer as valuable on the base paths (and this article suggests that that fact might be up for debate), but just keeps getting better with his bat to offset the decline. If nothing else, I think this article serves as a nice reminder on the merits (or lack thereof) of small sample sizes.

Dave Cameron writes about pitch framing and how it may or may not impact his vote for NL MVP. While I see the merits of pitch-framing and have no trouble seeing it's value to a team. I'm not sure if that skill alone is worth considering a player for the MVP ballot. That said, I love Cameron's willingness to entertain all options and really put some thought into something like this. Defense is part of the game too, and preventing runs is probably just as important as creating them.

Magic: The Gathering

Reid Duke explaining What is Magic? I think this is both a perfect primer for a new player to the game, as well as a good way of explaining what the game is for people who just don't understand. 

Marshall Sutcliffe explaining the his and Brian Wong's "Quadrant Theory" for drafting. I think like it a lot more than BREAD. The quadrant theory makes you think more about a card than "is it good", which is all I have found myself doing when I consider BREAD while drafting. The quadrant theory makes you consider WHEN a card is good. Which I think will lead to more synergy throughout a draft deck and, therefore, a better draft deck.

Mike Linnemann explaining some of the real-world inspiration for Khans of Tarkir. If you like to dabble in being Vorthos (as I do), this is the article for you. 

Nick Vigabool shows some statistics on how M15 has impacted standard. As for me, M15 has led to many different deck ideas (with varying levels of success). My favorite of which was a mono-red deck featuring Goblin Rabblemaster, which I was delighted to see pop-up at Pro Tour M15. My deck is built slightly different from the one Jeremy Dezani and the like ran, but that is because I have set mine up for post-rotation. I've had quite a bit of success with it nonetheless. Here is a decklist if you're interested.

Miscellaneous (a.k.a. reading about Batman)

I spent a lot of time this week looking at Comics Alliances's posts on Scott Snyder and the Batman: Zero Year story arc. There were several different interviews with Snyder that took place in different times throughout the arc, but my favorite by far was the last interview that took place after the story was over. It gave me a whole new outlook on the story, and gave me a better appreciation of both the story itself, and the choices the creative team made.

I also enjoyed this post from Andrew Wheeler on the story from the perspective of a new reader. I don't really have anything to say about it, I just like to read new and different perspectives on things I've been involved with for a long time. 

8.22.2014

Luyman's Term's #12: M15 Limited Report

This past July, I attended my first ever Magic pre-release event. Since it was my first, I set a simple goal for myself of not leaving with a losing record. I walked in with a decent idea of how I would accomplish this. I thought that a WU aggressive would perform pretty well, so I thought I would try that. To give myself the best chance at this, I decided to pick blue as my color because I felt it was stronger than white overall, and I wanted to be able to have a decent back up plan in case WU didn't work out. I thought that blue paired with any other color would be stronger than white paired with any other color. Also, I really wanted to run Ensoul Artifact on an Ornithopter for the dream 5/5 flier on turn 2.

Well, What Did You Open?

I opened my packs and immediately went for the rares. I found myself looking at the following:
  1. Ob Nixilis, Unshackled
  2. Sliver Hive
  3. Sliver Hivelord
  4. Chasm Skulker
  5. Shield of the Avatar
  6. Mercurial Pretender
  7. Ob Nixilis, Unshackled II
I found myself slightly underwhelmed, as I was hoping that I would open at least 3 rares in either white or blue. I decided all was not lost, as I still had a slew of common and uncommon cards to look through. I went ahead and sorted everything by color, and counted how many cards were in each pile. I don't remember exactly how many cards I had in each color, but I do remember my white cards numbered in the single digits. There was absolutely no way I was going to be able to run my dream WU deck. Once again, I found myself slightly underwhelmed. My black looked to be pretty good though, so I decided to go UB.

What Did You Build?

Here is the deck I ended up running:
While my dream of a WU aggro deck didn't quite pan out, I would have the opportunity for the Ensoul/Ornithopter dream to come to fruition and I was fine with that. I thought I might be able to steal a game or two with that combo (and that I did). I felt pretty confident in my deck, thinking that my main win-con would be just to stay alive until I could make Chasm Skulker/Ornithopter huge, or make things unblockable with the Pathmage. 

While I don't remember exactly how each of my rounds played out (in my defense, this was 6 weeks ago), I do remember how the deck performed overall...which was underwhelmingly. It was inconsistent at best, and a typical newby build at worst. I'm almost certain I fell into the trap of playing as many rares as possible, as Mercutial Pretender was almost always a 6-drop Typhoid Rat. I'm sure I could have put just about anything else in that spot and been better off.

How Did You Do?

I ended up going 2-2-1 in matches and 8-6-1 in games. Both matches I lost went to three games, so I consider the night a success. I've since participated in one other sealed deck event (going 2-2) and an M15 draft (2-1). I remember a lot less about these two events, but I do remember I ran WB in both and that I opened 2 Souls of Theros and a Liliana Vess in the sealed pool. So, overall I have a winning record in M15 limited, which is a lot more than I can say for JBT limited. I think I'm getting better, but the sample size is probably too small to know for sure. What do you think of my deck build? Do you think I fell into the trap of trying to play too many of my rares? What do you think were cards I could have sided in/out against decks I was weak against?

8.20.2014

Luyman's Terms #11: The Value of a Bat, Part 2

In part 1, I discussed the concept of run values, and how we can use them to quantify a players offensive value to his team. I concluded the post saying that Jermaine Dye being worth 0.168 runs with his bat was a good thing, despite looking less than impressive. So why is 0.168 runs a good thing? What if I told you that that the average MLB player was worth -0.777 runs with his bat over the course of the 2002 season?

I understand if that strikes you as counter-intuitive, if the average player was worth negative runs to his team how did anyone score? The answer is simple: while this metric scores an out as a negative event, an out does not result in runs (literally) being taken off of the scoreboard. This explains why a player can be worth negative runs to his team, even though a "negative run" is not a thing the scoreboard measures.

What follows is a list of all MLB players who ended the 2002 season with 300 or more plate appearances. Sheet 1 contains the raw number of times each even occurred for each player. Sheet 2 contains the number of runs each player produced with each event. The last column of sheet 2 contains the total number of runs produced by that player (note that all of the "value" columns measure value in runs):


I explained where these numbers came from in part 1, but it's been a couple months so I will explain again. Each "value" column is calculated by taking the count of a certain event and multiplying by how many runs that particular event is worth. Let's so a sample calculation using Jim Thome and singles:

Jim Thome hit into 73 singles in 2002
A single was worth .475 runs in 2002
73 singles * .475 runs = 34.675 runs generated by singles by Jim Thome in 2002

Add up all of the non-out values for each player and you get the "hit value" for each player. This number is the total number of runs generated by that player's hits.

Unfortunately, we cannot stop our analysis here. While score is kept by only counting up, the value that a player provides for his team is certainly not only positive. This is why we must also include the value of the outs a player generates as well. To account for the full value of a player, all we have to do is add his hit value and out value:

Jim Thome Hit Value = 162.777 runs
Jim Thome Out Value = -99.866 runs
Jim Thome Offensive Value = 162.777+ -99.866 = 62.911 runs

I started this piece by saying that Jermaine Dye's 0.168 runs of offensive value being a good thing, that number seems to the dwarfed by the offensive value generated by Jim Thome! In fact, there are quite a few players who dwarf the value put up by Jermaine Dye! How is 0.168 runs of offensive value a good thing? Well, even with the massive values put up by the top of our list, the players at the bottom of the list also played in the major leagues. We must also consider their contribution.

To find the average value of a player with his bat, all we have to do is add up the total offensive value for each player, and divide by the total number of players in the sample (or, in this case, let the spreadsheet do it for us). The total offensive value for the group was -210.508 runs, spread over all 271 players means that the average player was worth -0.777 runs to his team with his bat. If Barry Bonds is removed from the sample (which I will do, since he is such an outlier), the number goes all the way down to -1.169 runs.

Using the average player as benchmark, we can see that Jermaine Dye is ever so slightly above average. He actually ranks 105th out of the 270 players in the sample (remember, we are excluding Bonds)! What this tells me is that an average player is actually quite valuable, even if the average player doesn't seem so good on the surface.

Are there any burning questions you have about baseball that you would like for me to attempt to answer using math (poorly)? The next topic I will attempt to answer will be pertaining to pinch hitting in some form or fashion, but I would love to hear what you guys would like to see analysed!

8.18.2014

Link City #1

I'm back! Did you miss me? It has been a couple interesting months. After I posted the first part of my offensive values post, I realized that I had no idea what I was doing. I had just kind of jumped into the deep end and hoped I could swim. I didn't fully realize all the time I would need to fully research and understand the data I was messing with and then be able to flesh out an article that was interesting to read and I could feel proud of. My hope is that now I am comfortable enough with my data and that I have figured out how to better manage my time to I can both put out timely writings and have them be meaningful.

My goal for the foreseeable future is to post three times a week. I want to have a post full of interesting links (and maybe some commentary) on Mondays, a post on baseball/sabermetrics on Wednesdays, and a post about whatever (more than likely MTG) on Fridays.

Here are some things on the internet I have read in the last couple weeks that I have found interesting:

Baseball

First, is an article from Tom Verducci on the merits of banning the shift in baseball. While I can understand his position, he seems to focus quite a bit on one type of player and his stats seems to be a bit too cherry picked for my taste. While it is indisputable that offense is down as a whole across MLB, the shift is just a scapegoat I think.

Here is an article from Johnathan Judge as well as a quick post from Dave Cameron here explaining that Verducci's methodology may be over-simplified. They do a much better job of explaining why than I could.


Magic: The Gathering

I happened upon an interesting way to separate Magic cards from Reddit user u/alexpwalsh. I spent an hour or so constructing the separators, and I think they look great!

Over on Channel Fireball, Frank Karsten takes at look at "the optimal mana curve" for both constructed and limited decks using computer simulations. He even goes so far in one case to take a look at when you should mulligan your opening hand. It should be noted that his methods are completely context neutral, and are probably not super useful in practice. But it's certainly an interesting read, and definitely an interesting place to start if you're trying to build "the perfect deck."

Also on Channel Fireball, Owen Turtenwald tell us all about his Pro Tour M15 experience. Watching his reaction when he found out he made top-8 was something I will never forget. I hope some day that something will make me jump around like a kid again with a big grin on my face the way he did.

Miscellaneous?

I think that will be it for this week. Something quick just to get back into the swing of things!