DC United: The Good, The Bad, and The Playoff Situation

Discussion in 'D.C. United' started by Knave, Sep 10, 2012.

Thread Status:
Not open for further replies.
  1. DCUSA

    DCUSA Member+

    Jan 14, 2006
    Virginia
    Club:
    DC United
    Nat'l Team:
    United States
    Hopefully, because United has to go to Chicago for the last game and the Crew hosts TFC. United will need a little cushion there.
     
  2. MagpieFan

    MagpieFan Member+

    Apr 25, 2004
    Back in DC
    Club:
    DC United
    Nat'l Team:
    United States
    Knowing how "big games against Chicago" have gone historically. Yeah, let's sew it up before that.
     
    FLSoccerGran repped this.
  3. Knave

    Knave Member+

    May 25, 1999
    Eastern Conference
    GPGRWDLPTSPPGPRJMPPSE#WS#M#T#CD#AD#W%
    01SKC2806150607511.82626914-1310260.200.5764.3
    02CHI2806150508501.79616813-1211250.370.7462.5
    03NYR2806140707491.75606712-1112240.540.9062.5
    04HOU2905121007461.59546106-0815181.241.6958.6
    05DCU2806130510441.57536207-0617191.371.7455.4
    06CMB2806120610421.50516005-0419171.702.0753.6
    07MTL3004120315391.304451-04-0122083.303.8645.0
    08PHI2608070613271.043551-041134083.153.4338.5
    09NER2905070715280.973343-121033004.845.2936.2
    10TFC2905050717220.762637-181639-066.046.4929.3


    Western Conference
    GPGRWDLPTSPPGPRJMPPSE#WS#M#T#CD#AD#W%
    01SJE2806170506562.00687419-18-0638-2.29-1.9469.6
    02SEA2806130906481.71586611-100230-0.96-0.6062.5
    03RSL2905140411461.59546106-080425-0.75-0.3255.2
    04LAG2905140411461.59546106-080425-0.75-0.3255.2
    05VAN2905100712371.284352-030113161.051.4846.6
    06FCD3004090912361.204148-070214121.562.0945.0
    07CHV2707070713281.043549-061022132.032.3438.9
    08POR2806070714281.003446-091022102.372.7337.5
    09COL2905090218291.003444-110921082.653.0834.5


    GP = Games Played | GR = Games Remaining | PPG = Points Per Game
    W% = Winning Percentage = ((W+1/2D)/30)*100
    PRJ = Projected Points = PTS + (PPG * GR)
    MPP = Max Points Possible = PTS + (3 x GR)
    SE# = SS Elimination Number = MPP - (1st Place PTS) + 1
    M# = Magic Number = (11th Highest MPP) – PTS + 1
    T# = Tragic Number = MPP – (10th Highest PTS) + 1
    WS# = Wooden Spoon # = (18th Highest MPP) - PTS +1
    CD# = Calculated Despair Number
    Eastern Conference: (50th Percentile PRJ - PTS) / GR
    Western Conference: (44th Percentile PRJ - PTS) / GR

    The CD# means: a team must maintain a “#.##” PPG pace for the remainder of the season to end the season at the currently projected ##th percentile of their conference.

    AD# = Actual Despair Number = ([(Current 5th Place Points Per Game * 34) + 1] - Actual Points) / Games Remaining
    The AD# means: a team must maintain a “#.##” PPG pace for the remainder of the season to outpace and thus beat the current 5th place team's actual pace.

    - Sorted by PPG, then most GR.

    Strength of Schedule (factoring home and away), Home PPG, Away PPG
    SOSHPPGAPPG
    SKC1.88NYR2.54SKC1.77
    POR1.82SJE2.36SJE1.64
    FCD1.65CHI2.29LAG1.40
    NER1.62HOU2.29SEA1.33
    TFC1.51DCU2.27CHV1.31
    CMB1.51SEA2.15CHI1.29
    PHI1.45MTL2.13RSL1.07
    CHV1.43RSL2.07CMB1.07
    CHI1.42CMB2.00NYR1.07
    DCU1.42VAN1.92HOU0.93
    LAG1.41SKC1.87FCD0.79
    NYR1.38LAG1.79PHI0.77
    COL1.34POR1.67DCU0.77
    MTL1.32COL1.64VAN0.75
    SJE1.31FCD1.56TFC0.64
    SEA1.24NER1.53MTL0.47
    RSL1.14PHI1.31COL0.40
    VAN1.07TFC0.87NER0.36
    HOU0.91CHV0.79POR0.23


    Remaining Schedule
    WK29WK30WK31WK32WK33
    CHI: CMB@SKC PHI@NER DCU
    --------@NYR--------
    CHV:@CMB@DCU@VAN COL@FCD
    ---- RSL FCD--------
    CMB: CHV PHI SKC@DCU TFC
    @CHI----------------
    COL:@VAN LAG SJE@CHV HOU
    DCU:@PHI@POR@TFC CMB@CHI
    CHV----------------
    FCD:----@SJE@CHV@SEA CHV
    HOU:@PHI NER MTL PHI@COL
    LAG: TFC@COL RSL@SJE SEA
    MTL: SKC----@HOU@TFC NER
    NER: NYR@HOU@PHI CHI@MTL
    NYR: SKC TFC CHI SKC@PHI
    @NER----------------
    PHI: DCU@CMB@CHI@HOU@SKC
    HOU---- NER---- NYR
    POR:@SJE DCU@SEA@VAN SJE
    @RSL----------------
    RSL: POR@CHV@LAG@SEA VAN
    SEA: SJE@VAN POR RSL@LAG
    ------------ FCD----
    SJE: POR FCD@COL LAG@POR
    @SEA----------------
    SKC:@NYR CHI@CMB@NYR PHI
    @MTL----------------
    TFC:@LAG@NYR DCU MTL@CMB
    VAN: COL SEA CHV POR@RSL


    ---

    These new tables just aren't as good as the old ones. And formatting them with colors would be a lot of work.

    I abandoned my old Historical Despair Numbers because of the new conference system. (That said, those old anecdotal numbers will still probably be true for the league as a whole, just not in the Eastern Conference.) Instead I'm trying out a new Calculated Despair Number based (see above). In practice it may not be too different from the Actual Despair Number I've been looking at for years. They're pretty close right now -- basically an upper and lower minimum for playoff viability given current conditions.

    The Western Conference: FCD is still alive with that win, but VAN is in a much stronger position going forward. FCD has the 3rd hardest schedule in the league, VAN the 2nd easiest.

    The Eastern Conference: This is the only real playoff race and we're in it. First, MTL is toast. There's no way they're coming back. Even winning out might not be enough given current trends (see the two despair numbers over 3.00 PPG). As for DCU, that win last night was huge (even if it was lucky). And CMB's loss was just as big. The two teams are really pretty evenly matched going forward, DCU's schedule being just a hair easier than CMB's. The second to last game of the season -- CMB @ DCU -- may very well decide things.

    Even if we make the playoffs, it's very likely going to be at 5th place. That's because we're very, very unlikely to overtake HOU. HOU has the easiest schedule in the league.
     
    BBBulldog, greatscott, Winoman and 8 others repped this.
  4. Boloni86

    Boloni86 Member+

    Jun 7, 2000
    Baltimore
    Club:
    DC United
    Nat'l Team:
    Gibraltar
    Houston and Columbus both have 4 home games left while we only have 2. Our recent road record suggests we're probably not likely to get into playoffs. No better time than Thursday to break that trend
     
  5. Knave

    Knave Member+

    May 25, 1999
    That's true. Our away record is pitiful.

    (By the way, check out Chivas on their home and away records: .52 PPG better on the road than at home. This has been true of that team for years, IIRC.)
     
  6. Heist

    Heist Member+

    Jun 15, 2001
    Virginia
    Club:
    DC United
    Nat'l Team:
    United States
    The HDC isn't really their home. They've always been tenants, so those numbers are no surprise.
     
  7. spot

    spot Member+

    Nov 29, 1999
    Centennial
    Club:
    Colorado Rapids
    Nat'l Team:
    United States
    Great work. I used to think it was painful watching this week-to-week wondering if the Rapids would get into the playoffs. This is worse.
     
  8. Cweedchop

    Cweedchop Member+

    Mar 6, 2000
    Ellicott City, Md
    Columbus will most likely leapfrog United on Wednesday night when they host Chivas USA - obviously we are all pulling for a loss or a draw.

    United will have the opportunity to counter on Thursday. A draw on Thursday gets United ahead of Columbus regardless of Columbus's result (United has the goals scored tiebreaker in their favor).

    No better time than now to finally break their long drought on the road - not entirely hopefull, but the opportunity is there.

    Not to mention that if United win on Thursday :eek:, they'll jump over Houston and into 4th place (both teams will be tied on games played too)
     
  9. fischy

    fischy Member

    Aug 27, 2006
    My take -- Houston's got what should be at least 9 or 10 points left in their schedule. DC will have to do well to take 9 points, and we aren't likely to take 11 or 12, so catching Houston is unlikely. Chicago has pulled away, so we're not gonna catch them either.

    I do agree that the DC and Columbus results this weekend were huge for us, but especially so because I don't agree that we have an easier schedule than Columbus. You're talking about averages, but ya can't look at ti that way. You have to break it down, game by game, looking at each individually. Columbus has what should be 3 gimme games at home. That's 9 points. DC has to get 7 then, just to keep the lead -- actually it would be a tie, with DCU holding the tiebreaker advantage. With only one gimme game at home, that sets out a big challenge. It means beating Columbus -- and it means taking at least one point out of 3 road games. Really, we'll have to do better than that, because Columbus might take some points at home against SKC or at Chicago. So, while Columbus has 3 really tough games left, their 3 gimmes mean they have the easier schedule, for me. DC is goign to have to work much harder to get neededresults.
     
  10. Stan Collins

    Stan Collins Member+

    Feb 26, 1999
    Silver Spring, MD
    I disagree that you look at it game-by-game, because that introduces the bias of turning likelihoods into certainties, eliminating mean regression. I think your example is a case in point. Typically, when a team has three 'gimmes', that doesn't actually total 9 expected points. Suppose you have an 80% chance of winning each game--something that, in MLS, between parity and the possibility of a tie, is almost never true--the odds of winning three in a row are basically 50/50. Supposing they really are 'gimme' games, I would guess the actual expected points is probably somewhere closer to 6.
     
  11. tab5g

    tab5g Member+

    May 17, 2002
    Generally speaking, MLS is founded and operated on the principle/reality that there are, or there should be, no 'gimme' games.

    Not that it is impossible for CMB to take the full nine points from their home games vs CVS, PHI and TFC, but here are those three team's current away records:

    W-D-L
    4-5-4
    2-4-7
    2-3-9

    That's 20 losses (along with 12 draws and 8 wins) in 40 road games.

    While CMB is currently 8-2-3 at home.

    So I'd agree that (from those 3 specific home games for CMB) the actual expected points is probably somewhere closer to 6 (i.e. 4, 5, 6 or 7) and not really the 'gimme' 9 for CMB.
     
  12. fischy

    fischy Member

    Aug 27, 2006
    A smart guy once said there "Lies, Damn Lies, and Statistics." Columbus has taken 27 out of a possible 39 points at home. So, by that standard they would take 6 or 7 points. However that tells you next to nothing. It doesn't take into account the strength of the opposition -- and nor does your citation of the road records of the weak sisters Columbus has left on its home schedule. so, they're not even remotely predictive. Again, the opponent is meaningful...and there are weak sisters whenever there are teams that have half as many points as the top teams.

    Chivas has 4 wins on the road. They own their first 3 road games, and two of those were against Portland and Toronto. The only road game they've won since? Portland, again. (Gives me some hope for United's prospects in the Rose City). Philly's own 2 road wins -- against Chivas, and a freak upset of LA. How about the Union's 4 road draws? Against Dallas, which had scored 2 goals in only one of its preceding 9 games, and had been blanked in its previous 2 home games; DC United, in a game referee Mark Geiger stole from United to give Philly the draw; and draws in their last 2 games against NE and, wait for it, Toronto. TFC's road wins: over Montreal, when that team was struggling badly, and over New England (them again). Their draws, including Dallas and Houston, twice, are a little more impressive, but I wouldn't give them good odds of getting a point in Columbus, a city which is not in Texas.

    Also, Columbus now -- with Higuain and Arrietta is a vastly different team than even two months ago. -- so, their home record since they added Higuain would be more telling. That's a W-L-D 3-0-1 record, with the lone tie being against the Galaxy. And, the game before that, when Arrietta got his first start, they beat DC. Pretty impressive record -- virtually perfect. In fact, beginning with that win against DC, Columbus has lost only 2 games -- at RFK, before Higuain joined the team, and their last two games including a poor showing in New England, and a loss at NYRB, without Higuain. 6 wins, 2 draws and 3 losses (2 without Higuain). 20 out of a possible 33 points. Using statistical averages, we could expect 9 or even 10 points out of their last five games...unless we figure just the average of their games with Higuain, and then it's 20 out of a possible 27 points.

    Statistics are meaningless without context, especially if you're using them as a predictive model. I'm not relying on statistics because they don't tell the stories of those games. I'm looking at the teams and making reasonable predictions regarding the outcomes of the games based on the teams' relative quality, home field advantage and intangibles such as expected intensity -- which is a predictive model that is far more predictive of results than are statistics that are based on seasonal averages. For example -- if DC still had DeRo, I'd be more confident in their prospects on the road. His absence has to be factored in -- which your numbers don't do. Another example -- from baseball. If you looked at the Nats' road record, or the Braves' home record, no way do you come out with a sweep by the Braves. But, if you look at which team would be more intense and playing for their playoff lives -- or the injuries the Nats had (and the absence of Strasburg) -- then, that sweep becomes less improbable and even predictable.

    So, I don't give DC the advantage in the race with Columbus. I think the Crew are more likely to get through -- but I haven't given up hope. I also see that DC can get through by beating the Crew and snagging 3 or 4 points on the road, which is possible, notwithstanding DC's miserable road record up to this point, because as your strength of schedule reflects, DC will be playing some of MLS' lesser lights.
     
  13. Knave

    Knave Member+

    May 25, 1999
    An updated chart ...

    Rolling 5-Game PPG Averages: United and Columbus

    [​IMG]

    We got a little uptick with the win. Two losses in a row brought Columbus down to earth.
     
  14. tab5g

    tab5g Member+

    May 17, 2002
    or, it tells you next to everything.

    again, it's not impossible for Columbus to take 9 points from their home games vs CVS, PHI and TFC.

    but the "standard" that Columbus "would take 6 or 7 points" seems (and is) much more reasonable than the "standard" that Columbus "would take 9 points" from those three 'gimme' games.

    they certainly are "remotely predictive" -- if not more so.

    the general point is that MLS is a fairly well-balanced league (if one looks at the range of PPG for all teams in the league).

    The "weak sisters" aren't really that "weak" -- relatively speaking. I'll stand-by the statement that generally speaking, MLS is founded and operated on the principle/reality that there are, or there should be, no 'gimme' games.

    CLB is not a "top team." (Neither is DC.)

    http://www.mlssoccer.com/standings

    if CLB does indeed take care of business (and gather the required points to pass United in the standings by season's end), then certainly CLB will be more of a "top team" than DC would be -- but neither team has done "enough" to prove their final relative positions yet.

    there's still work to be done.

    add all of that up, and I think most overall analyses would more realistically be looking at 6 or 7 (and not 9) points for Columbus in their remaining home games against CVS, PHI and TFC.

    you seem to be relying on statistics as a basis for your analysis, and the specific context you want to place around which statistics are most meaningful, relevant and/or predictive.

    certainly, but the MLB analogy would have even been stronger if that "even predictable" scenario were available and presented last week (before the games were played) and not after the fact of the sweep.

    I agree with this general and broader analysis of the remaining full schedule for MLS teams, specifically DC and Columbus.
     
  15. YellowSpine

    YellowSpine Member

    May 7, 2009
    Portsmouth, VA
    Club:
    DC United
    Nat'l Team:
    United States
    At this point DC is in control of their future. There are pieces and players that have the ability to keep it that way. But given recent history of past couple years, this is where it gets pissed away and choke.

    Olsen has tinkered all season. It's man up time and play those who will get us the "W"s. Or time to start looking for a new direction. Will know in a little over a month.
     
  16. Knave

    Knave Member+

    May 25, 1999
    Eastern Conference

    GPGRWDLPTSPPGPRJMPPM#T#CD#AD#W%
    01SKC2905160607541.8663690724-0.130.1965.5
    02CHI2806150508501.79616811230.560.8262.5
    03NYR2905140708491.69576412190.871.1960.3
    04DCU2905140510471.62556214171.271.5956.9
    05HOU2905121007461.59546115161.471.7958.6
    06CMB2905130610451.55536016151.671.9955.2
    07MTL3004120315391.30445122063.593.9845.0
    08PHI2707070614271.00344834033.763.9937.0
    09NER2905070715280.97334333-025.075.3936.2
    10TFC2905050717220.76263739-086.276.5929.3


    GP = Games Played | GR = Games Remaining | PPG = Points Per Game
    W% = Winning Percentage = ((W+1/2D)/GP)*100
    PRJ = Projected Points = PTS + (PPG * GR)
    MPP = Max Points Possible = PTS + (3 x GR)
    SE# = SS Elimination Number = MPP - (1st Place PTS) + 1
    M# = Magic Number = (11th Highest MPP) – PTS + 1
    T# = Tragic Number = MPP – (10th Highest PTS) + 1
    WS# = Wooden Spoon # = (18th Highest MPP) - PTS +1
    CD# = Calculated Despair Number
    Eastern Conference: (50th Percentile PRJ - PTS) / GR
    Western Conference: (44th Percentile PRJ - PTS) / GR

    The CD# means: a team must maintain a “#.##” PPG pace for the remainder of the season to end the season at the currently projected ##th percentile of their conference.

    AD# = Actual Despair Number = ([(Current 5th Place Points Per Game * 34) + 1] - Actual Points) / Games Remaining
    The AD# means: a team must maintain a “#.##” PPG pace for the remainder of the season to outpace and thus beat the current 5th place team's actual pace.

    - Sorted by PPG, then most GR.

    Notes


    No big update, just a few notes. Basically, for once, a lot of things went right for DCU. A couple wins for us, a loss for NYR. It would have been nice if CMB lost, but you can't have everything.

    For the moment we've got NYR, HOU, and CMB all jumbled together right next to us vying for the 3rd, 4th, and 5th spots. I fully expect HOU to rocket back up the table. Their schedule (@PHI, NER, MTL. PHI, @COL) is the easiest in the league. But it could turn into a real fight between DCU, NYR and CMB.

    I can say this: we're going to need every point possible. If the current trends hold, after this week it's going to take even more points to make the playoffs in the Eastern Conference -- somewhere probably between 53 and 55 points. That's a point or two higher than I was thinking last week.

    We've got 5 games left. We'd be wise to win at least 3 of them.
     
    sitruc and Winoman repped this.
  17. DCUSA

    DCUSA Member+

    Jan 14, 2006
    Virginia
    Club:
    DC United
    Nat'l Team:
    United States
    I feel a classic RedBull choke job coming on. It all starts when they lose in NE.

    @NE, TFC, CHI, SKC, @PHI
     
  18. Knave

    Knave Member+

    May 25, 1999
    Could be ...

    WK29WK30WK31WK32WK33
    CMB:@CHI PHI SKC@DCU TFC
    DCU: CHV@POR@TFC CMB@CHI
    HOU:@PHI NER MTL PHI@COL
    NYR:@NER TFC CHI SKC@PHI

    There's only one "6-point" game in there -- CMB@DCU.
     
  19. John L

    John L Member+

    Sep 20, 2003
    Alexandria, VA
    One of our problems is that the other three teams in immediate contention with us (NYR, HOU, CMB) don't play each other anymore - What benefits us is that NYR plays SKC and CHI, the latter being good for us either way - CHI is only two points above NYR, and while they still have a game in hand, three of those games are SKC and NYR and CMB

    We finish wtih CMB and CHI - That could be either really great for us, or kill us off
     
  20. Heist

    Heist Member+

    Jun 15, 2001
    Virginia
    Club:
    DC United
    Nat'l Team:
    United States
    It is strange that somehow in the more difficult conference, most of the teams fighting for spots don't play each other a lot in the final 5 games. We do have Chicago and Columbus, but only at the end of the stretch... We really need to take 4-6 points of the weak Western teams to give ourselves a legit shot!
     
  21. tab5g

    tab5g Member+

    May 17, 2002
    And might as well be really wise about it and just win the next 3 games.

    And then use the final two regular season games to get a better seed and have more time to sell post-season tickets.
     
  22. tab5g

    tab5g Member+

    May 17, 2002
    If it wasn't obvious before, now would be a fine time to really start hoping TFC, PHI and NE can play spoiler and grab some points (except of course when DC travels up to play at TFC).
     
  23. Heist

    Heist Member+

    Jun 15, 2001
    Virginia
    Club:
    DC United
    Nat'l Team:
    United States
    Winning the next 3 games probably won't clinch anything. If we win 2 of 3 and then beat Columbus, that might clinch it (depending on Columbus dropping some additional points). I don't think we'll have a shot to beat out Houston unless they fall apart. They have a very easy schedule.
     
  24. tab5g

    tab5g Member+

    May 17, 2002
    Agreed. (Hypothetically getting to 56 pts with 2 weekends still remaining in the season likely won't clinch an Eastern spot at that time. But it seems almost assured that 56 pts would be enough at the end of the season to get a team an Eastern playoff spot. Although, I'm guessing United could find a way to miss the playoffs with 56 pts earned.)

    Winning the next three (though not vital) would give DC the best shot of not only qualifying, but give them a decent chance to grab the 3rd or even 2nd seed. And it would make draws (and not wins) "acceptable/helpful/required" results in the final two games.

    Having said that, 9 pts from the next 3 for DC is probably unrealistic, and United will very likely be needing 3-4 points vs CMB and at CHI to close out the season and grab a 4 or 5 seed in the East.

    (But there are some very real benefits to qualifying as that 3rd seed or better.)

    And as for Houston, they do have a relatively easy schedule in the league. But they also do have a CCL midweek game to contend with in Oct, which may or may not be a factor on what they can/will do in the league.
     
  25. Petworth Posse

    Feb 24, 2009
    Club:
    DC United
    Houston has the best schedule but we play some really weak teams too. This is going to be determined by who can play consistently well, and that has not been our historic strong suit over the last five years.
     
    Q*bert Jones III repped this.

Share This Page