Jump to content

Mis-Placement


Recommended Posts

Most teams have played 15+ games by now. Looking at 12U stats it appears that 25% of teams are not playing in the optimal division. Playing too high, or too low for their level. Either scenario has negative impact on players. They either don’t get to play at the level they deserve, or loose most of the games…
Wondering if PAHL placement process should be revisited and improved? Such large mis-placement should not be the norm?

Thoughts?

image.thumb.png.4554b0972ca93c3e789fd17578c4d183.png

Link to comment
Share on other sites

I don’t thing that placements are that far off.  Neither system is going to be perfect.  Team players change from season to season.  How could PAHL do a better job?  They already squeeze tons of data and games into a short amount of time.  As for myhockey, I believe that it does as best as it can with the games that are actually entered.  How many games aren’t even entered for teams ranked 250 or more? Those teams don’t usually care where they are ranked.  You’re splitting hairs!  But, good thread and questions.

Link to comment
Share on other sites

More placement games would definitely help to improve situation.

And I wouldnt be raising this question if not for such a significant placement miss. Based on  nimbers 11 out of 41 teams are not in the right division.

There are teams who lost nearly every game. For them, the placement couldnt be any further off, and it would be hard to make a case for these kids that its all aboit fun...

Regarding My Hockey ranking. To my knowledge this is the best publicly available source of stats. Surprisingly, it is up to date, and have game stats for all League games, scrimmages, plus tornaments for team way into 700 rank (all A Major teams). Much more reliable than PAHL standings for determining overall team level.

Link to comment
Share on other sites

Based on experience with my kids, PAHL actually does a decent job with placements. It’s not realistic to have that much data (15+ games) prior to placing teams. There will always be overlaps in ratings between teams at the top and bottom of each division. When there are significant misplacements, coaches/parents are usually a factor. If teams are losing every game, it’s likely they pushed to be placed in a higher division.

A few more placement games at the 12u level might help. Did the increase in games at the 10u level that was implemented a few years ago help? There is no decent player data coming out of ADM, so the extra games at that level should be making a difference. With a year of data on the books, it should be less of a factor at 12u.

PAHL always seems to be looking to volunteers for the placement committee. If you think the process is broken and have ideas for improvement, consider throwing your hat in the ring. 

  • Like 1
Link to comment
Share on other sites

I would also like to see another analysis later in the season.  One of the comments about stated that some teams have zero or only a win or two, how could they be placed correctly?  I say, for every winner, there is a loser.  Look at any divisions in any league. Top will have close to all wins and bottom close to all losses.  Still think PAHL goes a fairly good job and agree that (sometimes) when it doesn’t, parents or coaches had influence.

Link to comment
Share on other sites

3 minutes ago, PuckHead7 said:

I would also like to see another analysis later in the season.  One of the comments about stated that some teams have zero or only a win or two, how could they be placed correctly?  I say, for every winner, there is a loser.  Look at any divisions in any league. Top will have close to all wins and bottom close to all losses.  Still think PAHL goes a fairly good job and agree that (sometimes) when it doesn’t, parents or coaches had influence.

This is a good point. Goal differential is a better indicator...you can lose all your games, but if most of them are by a goal or two, someone would be complaining about this team killing everyone in the lower division.

Link to comment
Share on other sites

28 minutes ago, Saucey said:

This is a good point. Goal differential is a better indicator...you can lose all your games, but if most of them are by a goal or two, someone would be complaining about this team killing everyone in the lower division.

I remember my son's squirt team was near the bottom of their division but they never got blown out, they were in every game but only had a few wins.  Low and behold they made the championship and lost by only 1 goal.  

Link to comment
Share on other sites

7 minutes ago, Jack Handey said:

It's also true that some people who know about MHR will manipulate ratings.  

How? Not that I don’t believe you, I just don’t see a way.  You win or you lose. The score is the score.  If a score or a game isn’t entered, it can be.  If a score or game was entered incorrectly, it can be changed.

Link to comment
Share on other sites

1. Reporting only those scores that boost your rating.

2.a. Willfully omitting scores that lower your rating.

2.b. (Corollary to 2.a.) Showing up to play in a different jersey or under a different name so that the outcome cannot be recorded against you.

3. Disputing scores that lower your rating so that they are removed from your record.

4. Reporting games where the roster is significantly different.

5. Recording forfeits as real games.

6. Scheduling games against teams with a rating of more than +7 over yours.   

These are the most common.

Yes, the opposing team can make a claim against all of these, but most people don't know enough about MHR to do that.

 

 

Edited by Jack Handey
Link to comment
Share on other sites

2 hours ago, PuckHead7 said:

I would also like to see another analysis later in the season.  One of the comments about stated that some teams have zero or only a win or two, how could they be placed correctly?  I say, for every winner, there is a loser.  Look at any divisions in any league. Top will have close to all wins and bottom close to all losses.  Still think PAHL goes a fairly good job and agree that (sometimes) when it doesn’t, parents or coaches had influence.

The same analysis later on in the season will not be as good for highlightting shortcomings of the placement process, because teams are now limited (for the most part) to playing within the division. This will result in swinning the team ranking toward division average.

For example, Erie 502 could be easily compeeting in AA division, and have a great competitive season playing teams at their level. But due to placement into A Major Black, they will have to play lower skilled teams in Black division. Even if they win, their ranking will go down, not up. As a result, comparing based on ranking later on in the season will only confirm the claim that placement process is "working" as expected, which is not the case.

As for misuse of MHR, hard to beleive that too many teams would engage in number rigging. There are 32 possible pairs of eyes for each score entered. Abuse may happen, but likely not to the extend that would deem it unreliable. Not sure.

Link to comment
Share on other sites

50 minutes ago, lferg said:

The same analysis later on in the season will not be as good for highlightting shortcomings of the placement process, because teams are now limited (for the most part) to playing within the division. This will result in swinning the team ranking toward division average.

For example, Erie 502 could be easily compeeting in AA division, and have a great competitive season playing teams at their level. But due to placement into A Major Black, they will have to play lower skilled teams in Black division. Even if they win, their ranking will go down, not up. As a result, comparing based on ranking later on in the season will only confirm the claim that placement process is "working" as expected, which is not the case.

As for misuse of MHR, hard to beleive that too many teams would engage in number rigging. There are 32 possible pairs of eyes for each score entered. Abuse may happen, but likely not to the extend that would deem it unreliable. Not sure.

They asked to move up PAHL said no.

Link to comment
Share on other sites

1 hour ago, lferg said:

 

For example, Erie 502 could be easily compeeting in AA division, and have a great competitive season playing teams at their level. But due to placement into A Major Black, they will have to play lower skilled teams in Black division. Even if they win, their ranking will go down, not up. As a result, comparing based on ranking later on in the season will only confirm the claim that placement process is "working" as expected, which is not the case.

 

Not sure I agree with this.  If they are a AA team, then they should beat teams in A Major Black by 3 to 4 goals on average, which will then keep their MHR score around where it is.  If they want to move up in score, they should be the teams by more than 3 or 4 goals.  However, if they struggle against say NP and Youngstown or perhaps lose to them, then their score adjusts down appropriately.  

Link to comment
Share on other sites

4 hours ago, Jack Handey said:

@lferg can you do the same for 18u 16u 14u to see if it is consistently off?   Does PAHL get it right or "righter" as the kids get older?

You might struggle at 14u with what to do with AA Minor teams and where they should fit among the other divisions.  AA Minor is such a bad idea and maybe lferg can show it.

Sorry Jack, didn't planned to crunch numbers for the entire association. All you need to do is to get the list of teams from PAHL standing page, and add rankings from MyHockey site, then sort by ranking.

I am not stating that PAHL gets it wrong every time at every age group. I am simply gathering feedback and ideas on possible improvements to placement process where it make sense. 

So far, the facts show that placement could have been done better. Not perfect, but better. I think the reason 2 placement game model fails here is because we are looking at 40+ teams for this age group. If there would be 20 or less teams, 2 games and "prior history" might have been enough to get it right. But with 40+ teams in the mix, it is just not working.

The appeal process, is not working well either for 2 reasons. The coaches are limited by two key factors: first, "prior history" statistics may be available to PAHL, but not to coaches. With only 2 games behind, and no other info, coaches would be hesitant to make a call on the move (up or down). Too many unknowns at the time the appeal must happen. Second, even if they decide to appeal placement, they depend on agreement of other teams from upper division to scrimmage them.  And I know for a fact, that there were situations where teams just couldn't get anyone to agree to scrimmage them. In that case, the appeal has no ground, and team is stuck where it was placed. 

That part of the process can also be improved either by assigning mandatory additional placement scrimmage, or else how. Otherwise, it's really not an option, but more of a chance...

 

Link to comment
Share on other sites

Prior history should be known by the association of every team. Most of them collect this data upon registration so even the club jumpers data is known to the new association. All associations are responsible for providing the PAHL with their idea of where each of their own teams should be placed. A solid association would involve the coach to some extent. Initial placements are released in mid August along with the two game placement match-ups. If a team believes they aren't placed correctly, there is plenty of time to play more than the two assigned games.

As far as MHR, it simply can't be used in any form or fashion because the initial rankings aren't released until mid-September and as others have mentioned, the manipulation is far more common than you would think.

The PAHL has its' share of warts, but team placement is definitely NOT one of them. As someone suggested previously Lferg, I'm sure the placement committee would welcome all of the time and data that you can provide.

Link to comment
Share on other sites

Dazed, thank you for point of view. I don't think the confidence you have in PAHL placement would be equally shared by, for example, Erie 502 and Wheeling 507, or Southpoint 478 and Foxes 277.  These teams are so far apart in skill level, that it's way beyond the "grey" area. Playing each other would simply degrade the definition of having fun for either side.

You are a hockey parent, place yourself in their shoes. You have to drive your kid 3hr one way to the game with almost certain worst possible outcome, and serious goal differential...  Explain to these 35 kids and their parents that there is no issues with their placement...

As mentioned above in the thread, some of these teams have tried to change their placement. Apparently, there wasn't enough time, or information, or other controls in this process to make it happen.

I think as a first step it's important to recognize that current placement process needs improvement. Denying that in the presence of factual numerical evidence is not very productive. As for helping the PAHL placement committee to improve their processes, as I mentioned above, this very post and feedback received is your and my and other folks contribution.

As far as not using MHR, sorry, but you are not correct. The scores are submitted as of Aug 25, and include PAHL placement games. It is the most accurate publicly available source for determining team skill level. I can't speak for the entire site, but I am yet to find any discrepancy in reported games for Tier 2 12U PAHL group.

  • Confused 1
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...