Page 1 of 6 123 ... LastLast
Results 1 to 12 of 66
  1. New Source for Stats

    Good morning everybody! Hope everyone is still enjoying the elite show we were privileged to watch yesterday. That's going to be a fun replay to watch for years to come!

    I wanted to share a link to a Google Drive folder I created to share some of my work (link here). I plan to use this thread to give you a taste of what I'm doing. Today, I'll highlight the game summaries I create (link). These summaries don't hold the typical stats you can find anywhere. I figure you already have a place to go for those. I'll share a few examples of what I've already done, but your suggestions are welcome and appreciated (drop a message here or on Twitter [link]).

    The Ravens with Lamar have a very particular style (especially when winning). They normally grind the opponent down, especially in the 2nd half, so I like to look at snap counts and ToP by quarter. You can find this information on the first sheet ("ToP" [link]) of each summary spreadsheet. You'll also find snap counts broken down by play type (dropback, run, spike, kneel) for each team in every quarter. Here's the snap/ToP table for yesterday:

    Q Bal Snaps Cin Snaps Bal ToP Cin ToP
    1
    17
    9
    10:04
    04:56
    2
    11
    23
    04:32
    10:28
    3
    7
    25
    02:53
    12:07
    4
    11
    15
    06:20
    08:40
    H1
    28
    32
    14:36
    15:24
    H2
    18
    40
    09:13
    20:47
    Total
    46
    72
    23:49
    36:11
    Shared Google Folder with Ravens spreadsheets, nextGen charts, and more! Please share my content! (attribution to Twitter requested)

    Knight of the Kingdom of Perfect Play, Student of The Bill James School of Stamping Out Bullshit. Main Sources: PFR, particularly the Play Index; for cap stuff, RSR's Brian McFarland (secondary: OverTheCap, Spotrac)





  2. Re: New Source for Stats

    I also create summaries for both teams broken down by quarter and down (see the sheet "Play Types" [link]). I've shared the Bengals run stats by quarter elsewhere (link), and here are the Ravens numbers on first down:

    possession down type Snaps Yards YPP Success SuccessRate
    BLT
    1
    pass
    11
    133
    12.09
    8
    72.73%
    BLT
    1
    run
    11
    54
    4.91
    8
    72.73%
    BLT
    1
    kneel
    1
    -1
    -1.0
    0
    0.00%
    BLT
    1
    total
    23
    186
    8.09
    16
    69.57%
    Shared Google Folder with Ravens spreadsheets, nextGen charts, and more! Please share my content! (attribution to Twitter requested)

    Knight of the Kingdom of Perfect Play, Student of The Bill James School of Stamping Out Bullshit. Main Sources: PFR, particularly the Play Index; for cap stuff, RSR's Brian McFarland (secondary: OverTheCap, Spotrac)





  3. Re: New Source for Stats

    There are 4 other sheets I've included:
    1. "Playcalling" (link) lays out every offensive snap conseuctively. I originally created it after the Chiefs game to see how much time the Ravens wasted between snaps, but the visual representation of run/pass balance is also useful.
    2. "QH/PD" (link) holds tables with each Quarterback Hit and Pass Defensed the Ravens are credited with.
    3. "Snaps" (link) has a few different parts:
      • First, it shows the official snaps by quarter for each player, broken into tables by position. It also provides an idea of how many players at each position the Ravens used per snap in each of those quarters and the game as a whole.
      • You can also see the personnel by quarter, as well as various other summaries (e.g. how many times the Ravens had 2 RBs on the field; note that my personnel numbers don't deal well with plays with J.Hurst in as an eligible receiver or RG3 in the backfield)
      • Finally, I break down play results by personnel, both overall and by play type. You can see the Ravens were very strong from 13 personnel yesterday (18.14 YPP).
    4. The final sheet ("Individuals" [link]) breaks down individual player's snaps to show the proportion of their snaps that were called runs/dropbacks, as well as the YPP and any TDs scored on each (dropbacks include stats on completions and INTs as well). There are 4 tables:
      1. "Highest Proportion of Called Run Snaps" shows the players "Real" snaps (penalties, kneels, spikes, 2PT plays all removed) and percentage of those "Real" snaps that were called runs (sorted by called run %). RG3's only snap as a non-QB was a run, so he's at the top of the list.
      2. "Highest Yards Gained Per Play on Called Runs" sorts the players by YPC. The Ravens ran 7 times with Hollywood on the field yesterday, gaining 10.86 YPC and scoring 1 TD.
      3. "Highest Proportion of Dropback Snaps" shows the players "Real" snaps (penalties, kneels, spikes, 2PT plays all removed) and percentage of those "Real" snaps that were dropback (sorted by dropback %). Mark Andrews played 22 "Real" snaps, and the Ravens dropped back on 17 of them.
      4. "Highest Yards Gained Per Play on Dropbacks" sorts the players by YPDB. The Ravens dropped back 14 times with H.Hurst on the field yesterday, completing 11 passes for gaining 209 yards (14.93 YPDB), scoring 2 TDs, and throwing 1 INT.
    Shared Google Folder with Ravens spreadsheets, nextGen charts, and more! Please share my content! (attribution to Twitter requested)

    Knight of the Kingdom of Perfect Play, Student of The Bill James School of Stamping Out Bullshit. Main Sources: PFR, particularly the Play Index; for cap stuff, RSR's Brian McFarland (secondary: OverTheCap, Spotrac)





  4. Re: New Source for Stats

    I hope you find these sheets useful! Please ask questions if something is unclear! Suggestions are always welcome (can I lay the data out more clearly? Is there a stat you'd like to see that isn't available yet? Something I haven't thought of?)! Also, please share my work anywhere you see fit. I ask only that you drop a link to my Twitter (link) attributing the work when you do.

    Over the next few days, I'll update some of the season-long sheets I've made and share some samples here. I look forward to hearing your thoughts, and go Ravens!
    Shared Google Folder with Ravens spreadsheets, nextGen charts, and more! Please share my content! (attribution to Twitter requested)

    Knight of the Kingdom of Perfect Play, Student of The Bill James School of Stamping Out Bullshit. Main Sources: PFR, particularly the Play Index; for cap stuff, RSR's Brian McFarland (secondary: OverTheCap, Spotrac)





  5. #5

    Re: New Source for Stats

    This is a lot of data. Good stuff!
    Although Walsh's system of offense can compensate for lack of talent; however, defense is a different story. According to Walsh, talent on defense was essential and could not be compensated for. What did Walsh do in 1981? He acquired physical and talented players on defense.





  6. Re: New Source for Stats

    Finished updating another sheet today (link). This one breaks every offensive snap into one of eleven types: pass, sack, scramble, spike, run, kneel, fumbled snap, fumbled exchange, 2PT attempt, penalty, or pre-snap penalty.

    These types are then re-categorized: passes, sacks, and scrambles are dropbacks. Spikes and kneels are clock plays. Fumbled snaps, fumbled exchanges, and 2PT attempts are "other." There's a ton of information that we can glean from this breakdown, including a slightly more accurate idea of the run/pass breakdown. As expected, Madman Roman calls for runs and passes in damn near equal amounts.

    Team type Snaps Gain YPP Success SuccessRate TD Completions Percent of Plays
    Ravens
    Dropbacks
    309
    2297
    7.43
    161
    52.10%
    17
    177
    52.64%
    Ravens
    Called Runs
    278
    1453
    5.23
    155
    55.76%
    14
    0
    47.36%
    Ravens
    Real
    587
    3750
    6.39
    316
    53.83%
    31
    177
    100.00%

    There's plenty of other stuff in the sheet, including a full summary for each game and the season. You can see what the average Ravens game has looked like (~72 snaps, ~34 dropbacks, ~31 called runs), or see snap totals by game and compare them to Football Outsiders and PFR (doing so will allow you to understand what each considers to be a "snap").

    As always, if you have questions, comments, or suggestions, you can post here or contact me on Twitter (link). I've got 5 or so more of these to update that I'd like to share with you all. Hope to get them ready in the next day or so. Then I'd love to take some requests if you've got 'em!
    Shared Google Folder with Ravens spreadsheets, nextGen charts, and more! Please share my content! (attribution to Twitter requested)

    Knight of the Kingdom of Perfect Play, Student of The Bill James School of Stamping Out Bullshit. Main Sources: PFR, particularly the Play Index; for cap stuff, RSR's Brian McFarland (secondary: OverTheCap, Spotrac)





  7. #7
    Join Date
    Jan 2019
    Location
    Boston, MA
    Posts
    201

    Re: New Source for Stats

    These are truly amazing. I showed some of your game summaries (not even the whole Drive folder!) to a non-Ravens fan friend of mine and he was blown away.

    Thank you so much for all the effort, Ravens Flock is lucky to have you





  8. #8
    Join Date
    Mar 2017
    Location
    near Asheville, NC
    Posts
    24,781

    Re: New Source for Stats






  9. #9

    Re: New Source for Stats

    This is dope.





  10. #10

    Re: New Source for Stats

    OrganizedChaos21
    Great stuff. Appreciate the work being put in. Regarding snap count I noticed Mark Andrews played the fewest snaps out of all the TEs. Is this a trend for him?





  11. #11
    Join Date
    Jan 2019
    Location
    Boston, MA
    Posts
    201

    Re: New Source for Stats

    Quote Originally Posted by pslholder96 View Post
    OrganizedChaos21
    Great stuff. Appreciate the work being put in. Regarding snap count I noticed Mark Andrews played the fewest snaps out of all the TEs. Is this a trend for him?
    Not OC21 but just looking at his numbers: In general, Andrews plays more snaps than Hurst but less than Boyle. Boyle has led TE snaps in every game this season, tied for first in two of them.

    Andrews has had the fewest TE snaps for the last two weeks (Cincy2, NE), as well as week 1. I speculate it's a combination of dominating leads to him not being called upon as much (Miami, Cincy2), as well as maybe Hollywood being targeted more in those games?





  12. Re: New Source for Stats

    Thanks for the kind words, everyone. I have a few more sheets updated and ready to share. Hoping to get to them by the end of the day.

    Quote Originally Posted by pslholder96 View Post
    OrganizedChaos21
    Great stuff. Appreciate the work being put in. Regarding snap count I noticed Mark Andrews played the fewest snaps out of all the TEs. Is this a trend for him?
    Quote Originally Posted by BhndEnemyLines20 View Post
    Not OC21 but just looking at his numbers: In general, Andrews plays more snaps than Hurst but less than Boyle. Boyle has led TE snaps in every game this season, tied for first in two of them.

    Andrews has had the fewest TE snaps for the last two weeks (Cincy2, NE), as well as week 1. I speculate it's a combination of dominating leads to him not being called upon as much (Miami, Cincy2), as well as maybe Hollywood being targeted more in those games?
    Andrews is used in very specific situations. I need to update the numbers, but through the bye, the Ravens dropped back to pass on ~75% of his snaps. Compare that to Boyle (38%) and Hurst (51%), and I think we can see why Andrews isn't getting as many snaps. When the Ravens want to run the ball (which they do just shy of 50% of the time), they largely don't put Andrews on the field.

    It's still been rare that Hurst outsnaps Andrews. As BEL20, Sunday marked only the 3rd time it's happened this season. Against Miami, Hurst got the majority of his snaps in H2 after the game was over. Against NE, Hurst got more snaps than Andrews in Q when the Ravens were working to run out the clock. Sunday was a similar story: completely equal use for Andrews and Hurst until Q4 (where Hurst had 3 extra snaps).

    Date Boyle Andrews Hurst
    2019.09.08
    43
    31
    43
    2019.09.15
    47
    42
    32
    2019.09.22
    43
    43
    22
    2019.09.29
    41
    29
    15
    2019.10.06
    51
    41
    32
    2019.10.13
    53
    37
    32
    2019.10.20
    39
    33
    26
    2019.11.03
    56
    23
    27
    2019.11.10
    35
    23
    26

    Score effects would be a good direction to investigate. I haven't yet done it explicitly. I'm not sure Hollywood's absence matters. The games Hollywood missed (10.13, 10.20, half of 10.06) look similar to 09.15 (09.22 and 09.29 aren't good comparisons imo because they were down big late in the game and threw nearly 100% of the time in Q4; then again, they inexplicably threw quite a bit against ARZ too).

    Something interesting that I'm just seeing now: it looks like Hurst is recently getting a significant share of Q1 snaps. That says to me that Roman is scripting him into the offense more. I'll need to dig deeper to see if this is a recent trend, but it might mean that Hurst is finally coming along. Or Roman isn't terribly comfortable with Andrews as a run blocker, and he values disguising his play call more than the difference between Andrews and Hurst. Or maybe he just wanted to attack the Pats and Bengals with their TEs; later today I'll post the numbers from 13 personnel on Sunday (they were very, very good).
    Shared Google Folder with Ravens spreadsheets, nextGen charts, and more! Please share my content! (attribution to Twitter requested)

    Knight of the Kingdom of Perfect Play, Student of The Bill James School of Stamping Out Bullshit. Main Sources: PFR, particularly the Play Index; for cap stuff, RSR's Brian McFarland (secondary: OverTheCap, Spotrac)





Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Link To Mobile Site
var infolinks_pid = 3297965; var infolinks_wsid = 0; //—->