Fighting Blind: Or How I Learned to Love (Fantasy) Sports

It shouldn't be a surprise to anyone in the UPL leagues that I have no idea what I'm doing. Its fine, most of the people playing fantasy sports can't really claim to know what they're doing either. What sets me apart is that I'm actually using fantasy sports to learn how the actual sport works.

Roland has been telling me for years that I needed to learn about sports. Mostly because he was tired of looking at the blank stare on my face while he gushed out about bad calls and bad choices that meant absolutely nothing to me. But he was also looking out for me in a way, because he was tired of me being useless in conversation and at that point in my life so was I. But I'm a lazy punk and just didn't have it in me to follow sports on the day-to-day basis - I needed something to be working towards.

So when the opportunity arose to jump in on the UPL baseball league last summer, I jumped on it. And hey, it turned out to be a great thing! I was able to carry on a conversation about sports! And I gave enough of a shit to keep the conversation going. Then the opportunity came to jump in on the UPL football league. I did some research, but still didn't understand the funciton of the Tight End well enough to realize that I probably didn't need to draft two of them early. Oops. And yeah, Vince Young as my Quarterback? Utter failure there.

Then somehow I won the fantasy football league, thwarting a Roland Slam.

I have no idea what I'm doing, for the most part.

But I might just be figuring it out.

M-V-P! M-V-P!

Okay. We're going to switch gears real quick and talk a little bit about basketball. While there's a real-life debate about who the MVP, the question of who the UPL MVP is just as pressing (at least, in our little world).

Now, if you follow the voting criteria that is most commonly use in real-life, you'd go with the best player on the best team, if that player is having an MVP-type season. If no one on the best team stands out, then you go to someone who has had a spectacular season, with some adjustment for team success. This year, the O.N. Thugs (please hold the applause until the end) have been on cruise control for the last 25% of the season, so the traditional vote could go to Steve Nash, Deron Williams, Paul Pierce, or Hedo Turkoglu.

Nash/Deron are 17/11 and 19/10.5, respectively. Both shoot over 50% from the field. Deron's 80% from the line, and Nash is 90%. Nash gets more 3's and a few more rebounds. Deron gets more steals and blocks, and both are over 3.0 A/T. Two very solid seasons.

Interesting, Turkoglu and Pierce have very, very similar stats, and have played similar roles for their teams. It's almost eerie, actually. Pierce/Turkoglu are 20/5/4.5 and 20/6/5, respectively, and are both right around 45% from the field and 83% from the line. Pierce gets a few more steals and blocks, and Turkoglu gets a few more rebounds and assists. Even their A/T's are just a touch under 1.65.

But this year, it would seem odd to have one of those guys as the UPL MVP, since a) they would split the vote, and more importantly, b), none of them are clear-cut top 5 (or even top 10, and perhaps even top 15) players.

This season, the real-life MVP talk has revolved around Kobe, CP3, KG, and LeBron. There have also been some minor rumblings for T-Mac, Amare, Manu, and others, which seems somewhat misguided this year. However, if Amare plays for a full season like he has since Shaq came over in the trade, he'll be right up there in the MVP discussion, since he'll be something like 30 and 10, on 60% shooting, and over 80% from the line. So what about in the UPL? We'll start with the same players, and take a closer look. And just for kicks, we'll put in Marcus Camby, though we won't really discuss him until the end.

For reference, we'll need some baseline stats from the league.

	GP	FG%	FT%	3PM	PTS
2nd Place	81	0.479	0.805	842	14320
Median	81	0.4675	0.7775	723	11922
Kobe	81	0.460	0.838	149	2303
CP3	78	0.485	0.851	89	1642
LeBron	74	0.483	0.715	112	2223
KG	70	0.540	0.801	0	1329
Camby	78	0.451	0.708	6	717

	OREB	REB	AST	STL	BLK	A/T
2nd Place	1331	5271	3117	921	639	1.870
Median	1239.5	4739.5	2570	781.5	549.5	1.695
Kobe	93	514	436	150	40	1.710
CP3	64	314	903	213	3	4.631
LeBron	132	588	534	136	80	2.136
KG	135	651	243	98	86	1.761
Camby	229	1029	258	84	241	2.186

Now, when you get into the discussion, we should look at the individual impacts, so for all cumulative stats, you can divide the team numbers by 10, since we have 10 starting positions. Note that this is not quite the entire season (probably based on an average of 81 games played, not 82), so you really should weight everything by the number of games played, but this is close enough for this discussion. Now, we can get into the difference, compared to an average starter that each player contributes.

	GP	FG%	FT%	3PM	PTS
Median/10	81	0.468	0.778	72.3	1192.2
Kobe	0	-0.00075	0.00605	76.7	1110.8
CP3	-3	0.00175	0.00735	16.7	449.8
LeBron	-7	0.00155	-0.00625	39.7	1030.8
KG	-11	0.00725	0.00235	-72.3	136.8
Camby	-3	-0.00165	-0.00695	-66.3	-475.2

	OREB	REB	AST	STL	BLK	A/T
Median/10	123.95	473.95	257	78.15	54.95	1.695
Kobe	-30.95	40.05	179	71.85	-14.95	0.0015
CP3	-59.95	-159.95	646	134.85	-51.95	0.2936
LeBron	8.05	114.05	277	57.85	25.05	0.0441
KG	11.05	177.05	-14	19.85	31.05	0.0066
Camby	105.05	555.05	1	5.85	186.05	0.0491

Okay. Now we're talking. First of all, this lets you see what each players adds, compared an average starting player on an average team, for each category. I'm not using position-specific numbers, so this is a bit general, and can be refined in a number of ways. Where things get really interesting is with the magnitude of difference in some categories. When you have a deficiency in a category, you have to rely on specialists. Having players that contribute heavily to a category such that you don't have to rely on specialists makes life much easier. These tables show the number of average number of players that each player's "excess" makes up for in each category. The first column is just the sum of all the numbers, and doesn't carry a lot of meaning, at a theoretical level.

	Total	FG%	FT%	3PM	PTS
Median/10	0	0.468	0.778	72.3	1192.2
Kobe	3.171	-	-	1.0609	0.9317
CP3	3.081	-	-	0.2310	0.3773
LeBron	3.993	-	-	0.5491	0.8646
KG	0.342	-	-	-1.0000	0.1147
Camby	4.168	-	-	-0.9170	-0.3986

	OREB	REB	AST	STL	BLK	A/T
Median/10	123.95	473.95	257	78.15	54.95	1.695
Kobe	-0.2497	0.0845	0.6965	0.9194	-0.2721	-
CP3	-0.4837	-0.3375	2.5136	1.7255	-0.9454	-
LeBron	0.0649	0.2406	1.0778	0.7402	0.4559	-
KG	0.0891	0.3736	-0.0545	0.2540	0.5651	-
Camby	0.8475	1.1711	0.0039	0.0749	3.3858	-

Look at scoring. Kobe and LeBron score so well that it's like having an additional player. What's amazing is that Kobe's 3PM are of the same magnitude. What does this mean? If you have Kobe, you can pick up a player that doesn't score at all, and won't lose much ground to the rest of the league, and may even pick up ground in 3PM. With rebounding, you see that KG is worth about 1/11th of a player in OREB, and about 3/8ths of a player in REB. Having KG lets you have weaker rebounders around you. LeBron also helps with rebounding, though he isn't strong as KG (1/15th and 1/4th of a player for OREB and REB, respectively). Moving to AST and STL, you see where CP3 really shines. With CP3, you can have 2.5 players w/ no assists, and still be in reasonable shape, and about 1.7 players w/ no steals. LeBron and Kobe are both good w/ AST and STL, as well.

At this point, I think that we can take KG out of the race, which isn't particularly surprising, espcially given that he's missed a number of games. And LeBron's numbers seem to indicate that he's superior to Kobe and CP3, we really need to look at the rate stats. First, you don't see much change in FG%, save for KG . Also, note that the FT% differences are a bit misleading, since you don't know the number of attempts. Relative to the median, Kobe actually has a greater impact on FT% than CP3, because he gets to the line 9 times a game, compared to 5 times a game for CP3. On the flip side, LeBron's FT penalty is actually greater than this measure indicates, since he's getting to the line 10.3 times per game. LeBron's FT% will penalize a team. The -0.625% that he shoots at, compared to average UPL starters, is probably more like a 1% penalty, given the high number of attempts that he gets. What does 1% less at the charity stripe get you? 1-2 points in league standings.

Using a similar argument, CP3's A/T ratio is actually a bit understated because it's weighted so much more from his high assist numbers. His benefit of .2936 is huge. If you take CP3 off of Westy's team, and replaced him with an average UPL starter, the team's A/T goes from 1.87 (11 points) to about 1.53 (5 points), and if you replaced CP3 with a PG w/ a 2.0 A/T and a relatively high number of assists, then you'd be looking at around 1.59 (6 points). CP3 was worth about 5 points in the league standings.

So at the end of the day, who was the UPL MVP? LeBron is beneficial in every single category, save FT%, which is sort of amazing to think about. And the way that he and Kobe scored realy lets their owners have flexibility in picking up specialists to round out their team. Kobe hurts you in OREB and BLK, not surprising for a SG, and every so slightly in FG%, again not surprising for a volume shooter, and actually helps you with REB, which is somewhat surprising, and his 3-point shooting is a huge factor. CP3 was dominating in both AST and A/T, and his steals (which weren't quite as hard to get as 3PM this year) really set the tone.

I think that LeBron and CP3 are neck and neck, with Kobe trailing ever so slightly, followed by Camby (though you can make a strong case for Camby, whose numbers in hard-to-get categories, i.e., BLK, OREB, STL, more than offset his poor shooting and scoring, though he runs into even more problems w/ his shooting %, since he's bad both from the field and from the line). So who's the MVP? One way to look at it is who would you draft first, if you knew the stats were going to work out this way? Personally, I would draft LeBron first, followed by Kobe, and then CP3 and Camby. Why? It's easier to craft a team with you don't have any particular holes to fill. LeBron and Kobe fit that bill, whereas CP3 requires that you be on the look out for rebounding and blocks pretty early on, and Camby requires that you find scoring and shooting. If you start a team with LeBron or Kobe, you will end up with smaller holes to fill, which is always an easier proposition that filling a big void.

So, this was a simple way to look at a pretty complex problem. The fun part for me is that last year, me and C-Lauff made a pre-draft trade, swapping our positions in the 1st and 2nd rounds. I got Kobe, and C-Lauff got CP3. Kobe was awesome, and CP3 was injured. This year, I was deliberating between CP3 and Steve Nash in the 1st round, and I went with the "safer" choice. I should have remembered that any player that is picked up the year after C-Lauff has them becomes incredible :-)

In any case, the leap that CP3 made this year was phenomenal, and he should be in the mix for NBA Most Improved Player, if you take that term literally. For the UPL, I think that LeBron (and maybe even Kobe, depending on your preference in designing a team) rank ahead of him, but it's close.

But MVP? Screw it. I'm going with Paul Pierce. Best player on the best team, which just happens to be mine :-)

-Chairman (aka O.N. Thugs)

Vantage Point

So why in the world would I not be 100% gung-ho over a chance to get a future HOF player like Albert Pujols? It's a matter of context.

Take a look at these numbers:

Scenario	Runs	HR	RBI	SB	OBP	SLG
Expect-Konerko	962	250	979	50	0.378	0.502
Low - Konerko	862	234	878	35	0.366	0.482
High - Konerko	1012	264	1020	85	0.384	0.512
Expect - Pujols	990	258	992	50	0.383	0.510
Low - Pujols	890	235	891	35	0.371	0.490
High - Pujols	1045	275	1070	85	0.389	0.520
2007 2nd	1008	237	1003	217	0.366	0.482
2007 Median	897	220	875	115	0.359	0.472
2008 Leader	968	236	963	213	0.364	0.481
2008 Median	843	207	823	109	0.355	0.465

Note that 2008 Leader and Median stats are discounted because of the increase from 12 teams, from 11 in 2007.

What you'll notice is that my team is slow. Real slow. I probably need to address that issue at some point. I'm not stealing any bases, w/ or w/out Pujols.

But more important to this discussion, you'll notice is that w/ Konerko, I was planning on doing pretty well on offense. The lines for expected values represent slightly conservative expectations. But the net result is that my team can mash, which is expected when you have a bunch of high-OBP power hitters (e.g., Teixeira, Utley, Aramis Ramirez, Dunn, Griffey, Burrell, and Konerko/Pujols). Additionally, my lineup also has a nice lefty/righty platoon situation with Chris Duncan and Troy Glaus, both playing for the Cards. So my projects really could be a little more aggressive, since those two are on the same team, I can maximize that platoon. And I've still got lottery tickets in Jason Giambi and Nick Johnson (more slow, high OBP, power sorts).

Now, my pitching is a completely different story. My bullpen should be good. I've got 2 legit closers in Joe Nathan and Francisco Cordero, and 2 potentially very good ones (depending on injury) in B.J. Ryan and Brad Lidge. You roll the dice there, anyway. But the problem is that my starting pitching is problematic.

My one reliable starter is Harang. After that? Pedro coming off injury (and leaving due to injury yesterday). Oliver Perez, who has had 2 good seasons and 3 terrible ones that took him out of the majors. Andrew Miller and the new look, 70-win Marlins. Edison Volquez, a 25-year old rookie pitching in one of the most hitter friendly parks in baseball. And some dude named Jair Jurrens (who doesn't even have a picture on the Yahoo! website - so I have no clue what he looks like). I've already fired Barry Zito. If you have Brett Myers at his 05/06 average as a starter, you expect something like 13-8, 200K in 210 IP, 3.81 ERA and 1.25 WHIP. That would stabilize my starting pitching quite a bit, and let me be more selective about when I started my unproven pitchers.

Overall, I think that I solidified my offense a little bit, but left my pitching staff in a somewhat precarious position. But, what I explained to C-Lauff was that I'd always have a nagging feeling if I turned down this trade, that I'd be afraid of being the Portland Trail Blazers passing on MJ because they already had Clyde Drexler...

...which leaves me wondering whether Jair Jurrens is black, white, brown, yellow, blue, or green.

-Chairman, aka O.N. Thugs

Starting Off With a "Wait... What?!?"

The season just started and people are ready to kill me.

I don't necessarily blame them. I mean, who trades off a first round pick before the year even starts?

The Jimmy Dix Longballs.

While a certain level of doubt might always exist during a trade, what surprised me was finding out that Chairman had his own doubts about the offer. My doubt came about once I admitted that I let a homeristic man-crush with a pitching upgrade on the side sway me into trading off one of the most powerful and effective hitters of the decade.

It was a balance of risk and reward - I could be getting two average guys for a hero, but the hero's arm might fall off at any point in the season. I chose to go for reliability at lower reward than a higher risk / higher reward situation. Or, as illustrated using their 2007 stats:

	H/AB	AVG	R	HR	RBI	SB	OBP	SLG
Pujols	185/565	0.327	99	32	103	2	0.429	0.568
Konerko	142/549	0.258	71	31	90	0	0.351	0.49
Loss		0.069	28	1	13	2	0.078	0.078
	IP	W	L	SV	K	ERA	WHIP
Myers	68.2	5	7	21	83	4.33	1.28

I still cringe a little bit whenever I look at the disparity of runs and on base percentage between Konerko and Pujols. But seeing the extra 83 strikeouts (at a rate of 1.227 K's / IP) and potential for wins and saves I felt a little better. If Myers were to become the closer for Philly - which he might be again after a medicre opening day start that was blown up by Tom Gordon - then the value of this trade leans in my favor. Was the numbers loss worth it?

It will always depend on Pujols. If his arm falls off then this becomes the trade of the year for just about everyone other than the ON Thugs.

-CJ

League Rules

Our leagues are a little different from traditional fantasy leagues, so a little bit of info can help frame any discussion that we have here. All of this is sort of mundane, so I wouldn't advise reading this. Instead, just refer back, in case you see a strange argument down the road.

One thing that we tried to do w/ the UPL leagues from the start was to make the game more balanced, trying to minimize luck as a factor. Obviously, you can't eliminate luck completely, but you can change things around so that luck is reduced.

For baseball, we originally started with a cumulative points league for the season. Baseball is the one sport where statistics are relatively well understood, and we wanted a game that reflected the long haul. A few of us had played on the old SmallWorld fantasy site (I'm not sure if they even exist anymore), so we were familiar with this method of accumulating points. The problem is that teams that fell behind would just quit on the league, which didn't make things fun at all. We also tried this for basketball, as well. Eventually, we went to a rotisserie-style league for both basketball and baseball. This let you take advantage of the long-haul mentality that characterize both sports, but still have a format where small differences matter, and no one's completely out of it until pretty far in the season.

Right off the bat, we used a 6x6 system in baseball, using a mix of traditional stats and the rate stats available on the Yahoo! site that we used. The result was that we used R, HR, RBI, SB, OBP, and SLG for our offensive categories, and W, L, K, SV, ERA, and WHIP for our pitching categories. We also went with deep rosters, going with 10 starting offensive positions (2 UTIL slots), 7 starting pitching positions (1 SP, 1 RP, and 5 P), and 22-man rosters, w/ 2 DL slots. We go the full 162 games on offense, and allow 1400 innings pitched. This league isn't built for casual play, since we tend to have at least 10 teams (12 this year), so managers are looking at 3rd and 4th starters, and maybe even set-up men for their pitching staff, and trying to round out rosters with guys who you normally don't see in shallower fantasy leagues.

Similarly, in basketball, we went more categories than you normally see: FG%, FT%, 3FG, PTS, OReb, Reb, AST, STL, BLK, A/TO. Essentially, you have steals and blocks standing along, and then you see overlaps between DReb and Reb, AST and A/TO, and in the scoring categories. So, you can see the give and take that often comes up. We also run deep rosters here, going to 14 man rosters (since the NBA did away with the Injured List a couple years ago), and starting 10 players (PG, SG, G, SF, PF, F, C, C, UTIL, UTIL). Again, you see managers having to make a number of pick-ups that you often don't see in shallower leagues.

With football, it's hard to completely remove luck, since you're going head-to-head. But, even there, we've tried to make things a little more balanced. First of all, we use fractional scoring. It mystifies me why some online leagues still don't use fractional scoring. Why should 19 yards be scored the same as 11 yards? Maybe you can make the case that it's easier to calculate scores, but when the computer is doing the calculation for you, that's just dumb. We also use bigger lineups. We start QB, RB, RB, WR, WR, TE, RB/WR, WR/TE, K, DEF, D, D, and have 16 man rosters.

We also play around with the scoring. You can't get around assigning 6 points for a TD, so we did that for all TD, including passing TD's. This increased the value of QB's in our league, for sure. But more importantly, what we did early on was to increase the value of yardage. Back in the day, the standard scoring used to be 20 rushing/receiving yards per point and 50 passing yards per point. We add on special teams stats, by rewarding kickoff and punt returns at 20yds/point. And started off with 10 yds/point for rushing/receiving and 25 yds/point for passing. Most leagues have changed to this method the last couple years (though we've had zero influence on this). Where we do change things around is by giving points for carries, receptions, and completions (and penalize for incomplete passes). We give 0.2 points per carry, 0.75 points per catch, and 1 point for each completion (and -0.75 for each incomplete pass). What this does is try to reflect the value of what happens between the red zones in a real game. We also penalize turnovers pretty heavily. Each turnover is worth 4 points, so an interception or a fumble lost can be costly. We've also added 2 individual defensive players the last couple years, which has been pretty interesting, actually. In this league, defense and special teams matter. The net result is that in our football leagues, we have a number of ways to win. Not everyone just goes RB, RB early. And WR is a huge factor, since giving points for receptions can really balance things out. That said it's still a football league, so luck matters.

Anyway, if you have survived reading to this point, you should probably ask yourself why.

-Chairman, aka O.N. Thugs