What we’re watching

Is Ian Desmond underrated?

With standard disclaimers about small sample sets and the general meaningless of Spring Training games, here are some stat lines we find interesting:

The Prospects:

• Mets 1B Ike Davis: 11-for-21 with 2 HR
• Pirates 3B Pedro Alvarez: 5-for-16 with 1 HR
• Rays 2B Sean Rodriguez: 7-for-14 with 3 HR and 1 SB
• Athletics 1B Chris Carter: 2 HR in 5 AB
• Nationals SS Ian Desmond: 7-for-14  with 1 HR and 2 SB

Positional Battles:

• Houston Catcher: JR Towles (8-for-12) vs. Jason Castro (6-for-12)
• Marlins corner positions: Jorge Cantu (2-for-13) vs. Gaby Sanchez (4-for-11) vs. Logan Morrison (1-for-14) vs. Rule 5 pick Jorge Jimenez (2-for-7)
• Yankees #5 starter: Phil Hughes (2 Inn, 1 ER) vs. Joba Chamberlain (1.1 Inn, 5 ER) vs. Sergio Mitre (5 Inn, 0 ER)

Coming off injuries or questionable seasons:

• Magglio Ordonez: 5-for-14 with 2 HR
• Conor Jackson: 6-for-12
• David Wright: 2 HR in 12 AB
• Rickie Weeks: 4-for-9 with 3 SB

The Market Report is a weekly analysis of player valuations in the fantasy marketplace in an effort to find undervalued commodities.

By Andrew Gephardt

Just as we did last time with Yahoo fantasy baseball rankings, let’s compare ESPN’s rankings to those of Mock Draft Central. Once again, only players ranked in ESPN’s top 200 were considered for the lists containing the widest spreads. After all, at that stage in the draft it becomes increasingly challenging to predict which players will be taken next anyway. Another thing to keep in mind is that ESPN standard leagues only use one catcher.

Here then are the greatest discrepancies between this pair of rankings:

10 Hitters Ranked Significantly Higher by ESPN (by Value Difference)

1. Brett Gardner
2. Lastings Milledge
3. Carl Crawford
4. Derek Jeter
5. Brendan Ryan
6. Juan Pierre
7. Delmon Young
8. Ichiro Suzuki
9. Franklin Gutierrez
10. Hideki Matsui

Continue Reading »

By Eriq

Last season, Joe Mauer took home the AL MVP after slugging 28 HR, 96 RBI, and a .365 AVG. It’s nearly indisputable that Mauer had one of the finest seasons ever by a catcher. Yet hardly any fantasy baseball idea is more controversial than the prospect of taking a catcher in the first round.

Some people who blanch at Mauer as a high pick might be nervous about the inherent risks of investing in a catcher, the one position on the diamond (other than pitcher) where no player can every day, and where injuries and wear are big concerns. Others point to Mauer’s HR-to-flyball leap (from 6.5% in 2008 to 20.4% in 2009), seeing it as a clear sign that “Mauer power” is illusory.

However, the best case for Mauer as a solid first-round pick has nothing to do with faith he’ll keep up the HRs. Nor does it have anything to do with batting average, his most famous advantage over other players. It comes down to another category in which he’s so spectacular, he holds the potential of helping a team go a long way toward winning it.

That category is runs — probably the least-considered one in fantasy baseball. It’s also the category that’s been shown repeatedly as most correlative with a fantasy team’s hitting success.

Competitors tend to dismiss the runs category, perhaps because they see it as outside of their control. A player crosses the plate, after all, not just based on his ability, but also due to other players’ ability to drive him in.

However, it follows that getting on base is a skill, and that those who do it well score more runs. Over the last three years, only three batters with more than 600 at-bats (Albert Pujols, Chipper Jones, and Todd Helton) have gotten on-base more than Joe Mauer.

The result is that Mauer crosses the plate a lot. Over the last three seasons, he’s averaged 85 runs scored. He scored 94 in 2009. In 2010, he’s projected to score 92 times.

How valuable is that? Let’s put it in perspective.

Among catchers, he’s a beast. Last year, he scored 30 more runs than the average Top-10 catcher. He did this despite missing the first month of the season. Here’s a graphical look at Mauer’s performance last year. Note, too, those 112 AB as a DH:

(Click the picture for a larger version.)

Going into 2010, Bloomberg projects Mauer to have a 25-run advantage on the top-10 leading run scorers at catcher. Is that a lot? Let’s look at how other players in the first round stack up in the runs category compared to the top-10 projected run scorers at their respective positions.

As you’ll see above, Mauer (+25) holds a bigger advantage compared to his peers than Pujols (+12), Chase Utley (+12), Alex Rodriguez (+12), and Ryan Braun (+1).

Perhaps it’s time we began to look at things differently. Maybe Joe Mauer is to runs as players like Carl Crawford and Jacoby Ellsbury are to steals. This may sound controversial, but after going through several drafts already this year, and then projecting the seasons out, I’ve typically found that the team in my league that drafts Mauer is usually the one projected to win the runs category — and give himself a huge leg up on a league title.

Cross-posted at bloombergsports.mlblogs.com . To check out Bloomberg Sports’ complete suite of fantasy tools, click here.

Tiering up: A dynamic approach to snake drafts

By Brian Mills

When competitors start doing research on the best players to take in fantasy baseball drafts, they’ll often come across notions such as, “This position is scarce.”

Scarcity refers to the total amount of talent at a given position.  The more talent, the less scarce one can find value at the position.  But how is this really relevant to our strategy?  If almost all catchers are terrible performers, is it better to target the few who outperform the rest, or wait and accept the bottom-feeders?

Here is where the concept of tiering comes into play. Tiers give us an idea of the number of players at a certain value level, before we see a significant drop in value at the position.  So, after you grab Joe Mauer, there’s a significant drop off in value at catcher to Brian McCann or Victor Martinez.  After those two, there’s a significant drop off to the next few catchers. This consideration might lead people to be more aggressive in taking one of the top three, knowing the ramifications for missing out.

The large differences in talent are often times what people refer to as scarcity, but that language is imprecise.  Scarcity is less important than the distribution of talent at a position.  By example, here we plot the 5×5 Roto talent put up by the Top 12 Outfielders vs. the Top 12 Catchers using ESPN’s projections for 2010:

As you can see, the slope is much steeper for the Catchers than the Outfielders at the left of the chart then flattens out to a bunch of equally untalented catchers.  Most of us know this happens, but it can always be instructive to look at a picture.

Mauer (#1 C) is not as talented as Braun (#1 OF) in terms of overall production in a 5×5 league; however, taking into account the steep drop to the next best catcher, it’s not completely clear whether it’s better to draft Braun ­or Mauer. Better overall value or a better positional edge?

Interestingly, this is a year where at every position stands a single player who is expected to widely outperform the next best player at the same position: Joe Mauer, Albert Pujols, Chase Utley, Hanley Ramirez, Alex Rodriguez and Ryan Braun.

Most people might look at these players and decide to dismiss positional weights, opting for Prince Albert. We caution against any quick conclusion. Instead, we set out to devise a way to measure variability in talent at a position so as to appreciate how one player’s edge at his position compares to another player’s edge.

Continue Reading »

A word cloud showing the concensus

By Eriq

Everyone has a slightly different definition of what constitutes a “sleeper” in fantasy baseball, but perhaps by aggregating the community opinion, we can see how the crowds are targeting those players who are expected to produce good value in 2010.

What you see below is the result of a massive data collection effort over the past month. In that time, we’ve culled thousands of player names in magazines, websites, blogs, Twitter who have been identified as sleepers heading into the upcoming season. We then transformed the raw data into one big word cloud: The bigger the name, the greater number of times that player has been called a “sleeper.”

There are two ways to view this:

• The glass half-full outlook: The wisdom of crowds has been shown to be very accurate. Last year, our 2009 sleeper word cloud correctly identified players such as Nelson Cruz, Pablo Sandoval, Jayson Werth, Adam Jones, Josh Johnson, and Tommy Hanson as players who generated great returns for their owners.
• The glass half-empty outlook: Nobody may actually be “sleeping” on the biggest names below. In this digital era, competitors are inundated with information and opinion and likely have seen these names touted as breakouts for the upcoming season. In other words, these players may not represent as great a draft value as some people suspect. Moreover, as the draft position on these players rise, so too does the risk. Fantasy competitors learned that the hard way last year on players like Chris Davis, Alexei Ramirez, and Alex Gordon.

(Click for a larger view.)

So there you have it.

The conventional wisdom for the 2010 season holds that sleepers for the 2010 baseball season include: Brett Anderson, Carlos Gonzalez, Jonathan Sanchez, Kelly Johnson, Julio Borbon, Brandon Wood, Drew Stubbs, Jorge De La Rosa, Martin Pradio, Homer Bailey, Howie Kendrick, Matt Thornton, Miguel Montero, David Price, Jeff Clement, Everth Cabrera, Jake Fox, Alcides Escobar, Sean Rodriguez, Ian Stewart, and Billy Butler.

We’ll check back at the end of the season to see how the masses did in their picks.

Digging Deeper evaluates the depth of individual positions for NL- and AL-only fantasy leagues.

By Ron Henry

We previously measured the pulse of 2010 catchers in a Market Report. Using Mock Draft Central ADP data, here were the resulting tiers:

1. Joe Mauer, Victor Martinez, and Brian McCann
2. Matt Wieters, Jorge Posada, Kurt Suzuki, and Russell Martin
3. Geovany Soto, Miguel Montero, Bengie Molina, Chris Iannetta, and Mike Napoli

I have bolded the players that are available in NL-only leagues and if there is one thing to take from the upper tiers this year, it is this: the general public believes in Brian McCann and is generally unsure about every other catching commodity.  This should be a big warning not to overpay or reach for any catcher in the NL unless he has a tomahawk on his chest.

Martin through Iannetta are currently falling within 18 picks of each other (69-87 ADP) and there is very little to differentiate them from each other.  But I’m here to say in shallower NL-only leagues (10 teams or less, 1 catcher only), there is no need to worry.  Because these five catchers do not offer much in high-end statistics, there are a few more catchers who should provide similar value later in the draft.

Let’s see who some of those options are.

Continue Reading »

Questions can be sent to editor AT fantasyballjunkie.com

By Eriq

Reader M.R. writes: “Just finished reading your impressive post on finding advantages using Yahoo fantasy baseball rankings and I’m trying to build a spreadsheet that compares ADP/MDP versus Yahoo towards revising my rankings. I have a couple of questions.”

• Is there value in ADP (average) vs MDP (median)?

Statisticians debate the merits of “average” versus “median.” They both have their advantages, and there probably isn’t a huge difference. We like median slightly better because it blunts the impact of outliers. The guy who drafts Rickie Weeks #1 overall as a joke won’t be included. It doesn’t make that much of a difference, however.

• Is there value in building composite of all three sources or is MDC sufficient?

Yes, there’s value there. Obviously, pre-rankings tend to influence how people draft. So what you see on one site might not be reflective of a draft held on another site. Mock Draft Central will give a quick-and-dirty look at drafts, but the more sources included, the more accurate one can measure expectations.

• Do you use the O-Rank from Yahoo’s player list or the actual default rankings in the pre-draft section?

It’s likely the same, although Yahoo may make some adjustments based on draft trends.

• Once you have the lower of Yahoo/ADP formulas, would your recommend adjusting to is as your pre-rank in Yahoo?

Negotiating a middle ground is probably the best course. It’s helpful to have a certain expectation of what others are doing because you don’t want to draft a guy in the 5th round if it’s highly likely others won’t begin considering that guy until the 9th round. On the other hand, you don’t want to get too cute. One thing people tend to miss is the variability of a player’s draft position. In sum, you want to be able to have your own rankings, but also have an eye on the community rankings for optimum draft performance.

The Market Report is a weekly analysis of player valuations in the fantasy marketplace in an effort to find undervalued commodities.

By Andrew Gephardt

Now that we’re into March, it’s time to really focus and plan appropriately for our drafts. Throughout the past few months we’ve analyzed some of the best values based on Mock Draft Central data. The vast majority of fantasy players, though, use other sites to run their leagues.

There’s no denying that the designated league site of play influences the results of any draft. Whether we like to admit it or not, most of us are reluctant to select a player far down the pre-set rankings list. Accordingly, let’s analyze the pre-set rankings so that we might have a better idea as to how our drafts will play out. First up this week will be Yahoo! Fantasy Sports.

In order to determine which players Yahoo values differently, let’s compare Yahoo rankings to those of Mock Draft Central. Before providing a list of the greatest discrepancies, however, a few short notes. First of all, we’re only including Yahoo’s top 200 in the list of players that the site is higher on than Mock Draft Central. After all, Yahoo’s rankings are only useful up to a certain point. Call it laziness or whatever you want, but I highly doubt the analysts at Yahoo actually think, for instance, that Chris Iannetta is only the 795th best player heading into 2010. Along those same lines, we’re only including players in Mock Draft Central’s top 200 in the list of players for whom Yahoo has provided a relatively low ranking.

With that out of the way, here are the players with the greatest discrepancies between the sets of rankings:

Continue Reading »

Napoli: Net Ability Player Optimization Lineup Index

By Dan Gardner

If you remember last week’s article, you know that I promised to demonstrate how one can figure the optimal lineup on auction day. Net Ability Player Optimization Lineup Index (NAPOLI) can provide the answer. It involves a fairly straightforward procedure – minimal thinking and a bit of organization will be involved. I’ll offer up an example lineup, a justification for the algorithm, and a methodology document if you’re not bored out of your mind.

Disclaimer: I am not a PhD

When attempting to buy the optimal, value-maximizing fantasy lineup, there are billions and billions of possible combinations to choose from. Consider a simple 10-team NL-only league that drafts C, 1B, 2B, 3B, SS, OF, OF, OF and 6 P – a total of 140 players will be drafted. Budget considerations notwithstanding, there are over 20 trillion possible ways to assemble your team. For larger leagues with more roster spots and “flex” positions, the number of combinations gets more ludicrous. How can one possibly sort through all these lineups?

Well, I’m not going to do that. I’m sure plenty of operations research PhD’s out there can solve this problem using some form of linear programming. I do not have those skills. What I do have is a sense of what their solutions will try to do, which is minimize the loss in efficiency from the profit-maximizing lineup. And I think I’ve come up with a straightforward method that sacrifices a little accuracy for a lot of understanding and ease of use.

In a Nutshell: Add MVPs to your Lineup

Here’s how NAPOLI works: Start with the most profitable lineup and then upgrade or downgrade as efficiently as possible to fit the budget. To identify opportunities to move closer to the budget without losing much profit, calculate the ratio of marginal Value$ added to marginal Market Price$ for each combination of players in the lineup and possible replacement candidates.

When the profit-max lineup comes in under budget, look for players with a high MV/MP ratio – as close to 1 as possible. When the profit-max lineup is over budget, look for players with a low MV/MP ratio – also as close to 1 as possible. The set of players with ratios close to 1 are MVP candidates. MVPs improve the lineup as much as possible when spending money, and damage it as little as possible when slashing spending.

Here’s a quick example. Just from eyeballing the CHONE spreadsheet used for our recent draft value chart, I calculated a good NAPOLI lineup for a standard Yahoo! league (C, 1B, 2B, 3B, SS, OF, OF, OF, UT, SP, SP, RP, RP, P, P, P). I used a $180 budget, which is a little over $11/player and in line with the traditional $260 for 23 players auction format.

With the exception of Beltre, Damon, Wells, Baker and Francisco, this was the profit-maximizing lineup. In order to get under-budget, I had to dump David Wright, BJ Upton, Carlos Beltran, Javier Vazquez and Carlos Marmol. I replaced them with the aforementioned 5 players, who offered similar profit at a reduced price. It’s a very pitching heavy team (over ½ the budget on arms!), but pitching seems to come cheap.

Continue Reading »

The Market Report is a weekly analysis of player valuations in the fantasy marketplace in an effort to find undervalued commodities.

By Andrew Gephardt

Continuing with our series on the best and worst values based on various projection systems, this week let’s take a look at Marcel projections. Respected sabermetrician Tom Tango developed this projection system. Marcel projections are very simple by nature, taking into account only the past three seasons of MLB statistics with recent data given the most weight.

Once again, keep in mind that the ADPs for superstars don’t correspond perfectly to their expected market values. No, Albert Pujols isn’t really overvalued. Then again, Jose Reyes and Troy Tulowitzki are indeed going too early in drafts based on Marcel projections.

You’ll notice that all of the best values among pitchers are starters, not relievers. Unlike CHONE, Marcel does project saves totals. However, the projections for saves are relatively conservative and in effect depress the dollar values of closers.

Comparing with last week’s analysis of CHONE, we do find some similarities. Once more, Russell Martin and David Ortiz rank among the best offensive values. Josh Beckett, Cole Hamels, Dan Haren, and Javier Vazquez similarly represent some of the best pitching values. On the other hand, Josh Hamilton, Aramis Ramirez, and Manny Ramirez are repeat offenders as the worst offensive values. Likewise, Daisuke Matsuzaka, Neftali Feliz, and Brandon Webb are among the worst pitching values.

At the same time, we also find a couple major differences between CHONE and Marcel. Heath Bell and Carlos Marmol were undervalued based on CHONE yet are overvalued according to Marcel. This reinforces the fact that the value of closers is largely tied to their saves total. Finally, one hitter was among the best values according to CHONE yet is the worst value overall based on Marcel: Jose Reyes is the ultimate boom or bust pick in 2010.

Here is a list of the best and worst values according to Marcel.

Continue Reading »