Interviewing/Bar Range Data Collection

[Ben E Lou]

I would imagine we might be able to tease out which staff members have the most impact on the range by posting some examples. This will be much easier for folks in my leagues than most, but all are welcome to participate if they want. Please use the format in the first post after this one.

EDIT: I AM LOOKING FOR PRE-INTERVIEW NUMBERS ONLY HERE.

EDIT2: I'VE GOT DATA FROM 60+ TEAMS NOW, AND A SPREADSHEET IS ATTACHED IF FURTHER ANALYSIS IS DESIRED.

[Ben E Lou]

INTERVIEWING RATINGS
HC: 97
OC: 77
DC: 28
AC: 83

RANGES
Offense: 22
Defense: 33
Special Teams: 20

[Ben E Lou]

INTERVIEWING RATINGS
HC: 68
OC: 29
DC: 72
AC: 76

RANGES
Offense: 33
Defense: 27
Special Teams: 24

[Ben E Lou]

INTERVIEWING RATINGS
HC: 67
OC: 43
DC: 37
AC: 61

RANGES
Offense: 32
Defense: 32
Special Teams: 27

[Ben E Lou]

INTERVIEWING RATINGS
HC: 82
OC: 41
DC: 63
AC: 55

RANGES
Offense: 31
Defense: 28
Special Teams: 29

[Ben E Lou]

INTERVIEWING RATINGS
HC: 76
OC: 42
DC: 51
AC: 67

RANGES
Offense: 33
Defense: 28
Special Teams: 26

[Ben E Lou]

Ok. That's the data for all of the leagues I'm in that have web sites easily reporting this information. My initial observations/theories on these five sets of data...

I'm guessing that no matter how bad your guys are an interviewing, 33 is the max.
It appears that the HC impacts offense/defense range, but not ST.
Coordinators matter, perhaps more than HC, for their respective sides of the ball.

[corbes]

INTERVIEWING RATINGS
HC: 93
OC: 59
DC: 65
AC: 72

RANGES
Offense: 26
Defense: 29
Special Teams: 27

[MIJB#19]

INTERVIEWING RATINGS
HC: 66
OC: 46
DC: 63
AC: 66

RANGES
Offense: 29
Defense: 24
Special Teams: 26

[MIJB#19]

Interesting, Corbes' staff is supposedly better across the board, but in practice my staff is better at defense and special teams.
Coincidentally, my staff members are top notch in 'scouting ability' though and aside from my head coach, better than Corbes' staff (assuming he used the date of IHOF's Toronto Lake Monsters).

[FootballDownUnder]

INTERVIEWING RATINGS
HC: 74
OC: 65
DC: 29
AC: 66

RANGES
Offense: 26
Defense: 38
Special Teams: 31

[Sea Kitten]

INTERVIEWING RATINGS
HC: 57
OC: 54
DC: 6
AC: 10

RANGES
Offense: 34
Defense: 45
Special Teams: 42

[Ben E Lou]

Heh. So much for the 33 theory. Need to look at those more closely later.

[Sea Kitten]

INTERVIEWING RATINGS
HC: 74
OC: 51
DC: 25
AC: 25

RANGES
Offense: 29
Defense: 40
Special Teams: 42

[corbes]

Yes that was Lake Monsters

[corbes]

CCFL:
INTERVIEWING RATINGS
HC: 74
OC: 65
DC: 90
AC: 71

RANGES
OFF 26
DEF 18 (actually varied between 18 and 20)
For DL, it was 19 across the board
For LB, it was between 18 and 20 depending on attribute
ST 23

Note that my 74 HC is the same as Sea Kittens in Post #14, with the 90 DC producing dramatically smaller ranges

[Gauss]

GML:
Interview Ratings
HC: 74
OC: 29
DC: 51
AC: 56

Ranges
Offense: 36
Defense: 34
Special Teams: 29

[Sea Kitten]

INTERVIEWING RATINGS
HC: 44
OC: 21
DC: 40
AC: 60

RANGES
Offense: 36
Defense: 37
Special Teams: 28

[Nemesis]

INTERVIEWING RATINGS
HC: 46
OC: 66
DC: 67
AC: 33

RANGES
Offense: 28
Defense: 29
Special Teams: 36

[Nemesis]

INTERVIEWING RATINGS
HC: 66
OC: 46
DC: 63
AC: 66

RANGES
Offense: 29
Defense: 24
Special Teams: 26

INTERVIEWING RATINGS
HC: 93
OC: 59
DC: 65
AC: 72

RANGES
Offense: 26
Defense: 29
Special Teams: 27


Veeeeedddy interrrrrresting.

[Ben E Lou]

Some early correlations make me wonder if the HC and AC's ratings are applied to one side of the ball depending on their positional orientation. The sample size of 13 data sets is small, so I wouldn't say anything definitive yet, but there is a decent indication that even AC does impact the range, but so far the correlations are

One correlation looks *very* strong, though: if anyone cares that much, the AC interviewing rating has a near-linear correlation with the ST range.

It just occurred to me that I might be able to grab significant amounts of data on this from my databases in one shot--like maybe close to 100 teams, depending on how many people uploaded their ratings bars. Let me see...

[Ben E Lou]

Ok. It's doable, but a bit tedious. Maybe some time this week. Until then, keep 'em coming.

[MIJB#19]

Veeeeedddy interrrrrresting.
Not really. As I mentioned above, the 'scouting ability' of the second staff is better than the first one.

[Sef0r]

Interview Ranges:
HC: 72
OC: 82
DC: 56
AC: 60

Ranges
Off: 22
Def: 32
ST: 33

Just because I don't know but are the ranges the same after a player is interviewed regardless of ratings? I.e. will the range be 11 for all?

[corbes]

In the Nemesis comparison:
the AC of the first team is defensive (LB)
the AC of the second team is offensive (QB)

[corbes]

HC of the first team is offensive (TE)
HC of the second team is offensive (OL)

[Nemesis]

In the Nemesis comparison:
the AC of the first team is defensive (LB)
the AC of the second team is offensive (QB)


Out of personal curiousity, what is the Scouting Rating for the HC, DC, and AC for both the first and second team?

[Ben E Lou]

I am going to pull scouting ratings, too, but indications still seem that interviewing is the biggest influence.

[corbes]

Don't have time to dig it out now but team #1 is IHOF Maassluis and team #2 is IHOF Toronto

[Yoda]

HC: 20
OC: 23
DC: 19
AC: 47

Offense: 38
Defense: 37
Special Teams: 38

[Ben E Lou]

Ok. I'm running some queries across four leagues and I have a few general data points.

1. Sea Kitten's TN team in the GML has the worst non-interviewed range I'm seeing: 45-46 on defense. His staff:
<table border="1" cellpadding="0" cellspacing="1"><tbody><tr><th>current_team</th><th>role</th><th>interviewing</th><th>scouting_ability</th><th>young_player_development</th><th>primary_group</th><th>Staff_ID</th></tr> <tr> <td>28</td><td>Head Coach</td><td>57</td><td>70</td><td>97</td><td>Offensive Linemen</td><td>285</td></tr> <tr> <td>28</td><td>Offensive Coordinator</td><td>54</td><td>34</td><td>83</td><td>Running Backs</td><td>251</td></tr> <tr> <td>28</td><td>Defensive Coordinator</td><td>6</td><td>41</td><td>68</td><td>Defensive Linemen</td><td>357</td></tr> <tr> <td>28</td><td>Assistant Coach</td><td>10</td><td>61</td><td>63</td><td>Tight Ends</td><td>447</td></tr></tbody></table>
The best non-interviewed range in those four leagues is 19, also in the GML, from barnrat's NJ team on offense. Their relevant data:
<table border="1" cellpadding="0" cellspacing="1"><tbody><tr><th>urrent_team</th><th>role</th><th>interviewing</th><th>scouting_ability</th><th>young_player_development</th><th>primary_group</th><th>Staff_ID</th></tr> <tr> <td>19</td><td>Head Coach</td><td>73</td><td>75</td><td>16</td><td>Defensive Linemen</td><td>5</td></tr> <tr> <td>19</td><td>Offensive Coordinator</td><td>91</td><td>85</td><td>18</td><td>Wide Receivers</td><td>405</td></tr> <tr> <td>19</td><td>Defensive Coordinator</td><td>75</td><td>55</td><td>56</td><td>Secondary</td><td>459</td></tr> <tr> <td>19</td><td>Assistant Coach</td><td>76</td><td>44</td><td>44</td><td>Quarterbacks</td><td>462</td></tr></tbody></table>
Because of the variations and because some players are interviewed, pulling usable data en masse isn't as easy as I'd hoped, but it's doable. There's no nice, neat flag in Conscriptor that says "this guy was interviewed," so instead I need to determine each team's most frequent differential on each side of the ball and use those numbers. Working on that now. My hope is to provide a big spreadsheet for analysis.
<table height="119" border="1" cellpadding="0" cellspacing="1" width="15"><tbody><tr><td valign="top">
</td><td valign="top">
</td><td valign="top">
</td><td valign="top">
</td><td valign="top">
</td><td valign="top">
</td><td valign="top">
</td><td valign="top">
</td></tr><tr><th>
</th><th>
</th><th>
</th><th>
</th><th>
</th><td valign="top">
</td><th>
</th><th>
</th></tr><tr><td>
</td><td>
</td><td>
</td><td>
</td><td>
</td><td valign="top">
</td><td>
</td><td>
</td></tr><tr><td>
</td><td>
</td><td>
</td><td>
</td><td>
</td><td valign="top">
</td><td>
</td><td>
</td></tr><tr><td>
</td><td>
</td><td>
</td><td>
</td><td>
</td><td valign="top">
</td><td>
</td><td>
</td></tr></tbody></table>

[Ben E Lou]

It appears that roughly 2/3 of owners in these four leagues upload bars to Conscriptor. RFOL's latest class was just released yesterday, and only 10 teams there have uploaded bars, so if I get this sorted out this morning, we'll have around 70 teams worth of data. That's probably enough to draw some conclusions.

[Shoveler]

FOOL
HC - 78
OC - 21
DC - 74
AC - 94

Ranges
Offense - 39-40
Defense - 22
Special Teams - 21

IHOF
HC - 28
OC - 49
DC - 56
AC -23

Offense - 30
Defense - 30
Special Teams - 35

GML
HC - 44
OC - 17
DC - 50
AC -47

Offense - 39-40
Defense - 27
Special Teams - 35

CCFL
HC - 75
OC - 65
DC - 69
AC -61

Offense - 30
Defense - 28
Special Teams - 26

[Ben E Lou]

Ok. I have data in a spreadsheet on the 68 teams for whom data is in Conscriptor. Here's what I am seeing.

1. The biggest influencers--by far--are the ones that I would expect based on the Help File: OC interviewing heavily influences offense range, DC interviewing heavily influences defense range, AC interviewing heavily influences P/K range. These correlations are so strong that it's quite possible that everything else is pretty much trivial.
2. There may be some noise, but it appears that the secondary influencers are AC Interviewing and the opposite Coordinator's interviewing.
3. HC Interviewing has little/no influence on range. (FWIW, by using drat class import, I've tested interview results, and it's clear that HC Interviewing has a *major* impact on the correctness of the OR/UR stuff.)
4. Scouting ratings have a small influence on ranges, but emphasis on small. It appears that there are tertiary.
5. MIJB did not upload his data, so the "controversial" numbers he is reporting cannot be confirmed independently. (MIJB, how did you get those ranges? Did you dig into the export files?) Generally speaking, they fit with what I'm seeing elsewhere, though.

[Ben E Lou]

Spreadsheet attached to first post if anyone wants to do further analysis.

[Ben E Lou]

Oh, and in some cases there was a small variation in the range, as some have reported. In those cases I used the highest number.

[MIJB#19]

5. MIJB did not upload his data, so the "controversial" numbers he is reporting cannot be confirmed independently. (MIJB, how did you get those ranges? Did you dig into the export files?) Generally speaking, they fit with what I'm seeing elsewhere, though.Correct that I didn't upload extractor results. I got those numbers directly out of the game. The scouted bars are exactly 100 pixels wide and the length of the blue bar is equal to the number that extractor would get. To verify this method should work, I cross referenced it with the staff bars and the numbers you gave for them. If you feel like that makes my data invalid, feel free to omit those numbers.