Login

Mediocre_Fred · 02-29-2020, 12:16 AM

[1,738 words, ready for grading]
[If possible, I'd like to use my first time media bonus, since I haven't actually been able to post anything since the season started due to being very busy with work]

Alright, fellas! At long last, I have something worthy of posting. A few weeks ago, I decided it would be rather fun to make a web scraper that would gather data on all of the SMJHL players. It was not a fun process like I thought it would be, but I stuck with it and gathered player data every week during the regular season. Every real life week, that is. There would be several sim weeks with no changes if I had decided to do it that way.

What I decided to do with all this data I gathered was a quick little comparison of the TPE between the different teams in the SMJHL. More specifically, I wanted to see how each team's average TPE changed week by week during the regular season leading up to the play-offs. My python script ran every week and would output a CSV file that contained all of the player stats found on each player's page. This includes TPE if it was listed, and if it was not listed, it was gathered by calculating the TPE based on each attribute's level. Something I had originally forgotten to account for was the different cap levels that are placed on SMJHL players. In order to calculate this capped TPE for each player, I would subtracted the 'Points Available' number from their total TPE numer. This ensured that every player would be represented how they would be "on ice". For example, a player may have 568 TPE total, but since they cannot use more than 425 TPE in the SMJHL, they would have 143 TPE saved in their 'Points Available'. 568 - 143 = 425 TPE, the amount that player would actually be able to use while playing for their team. The data is plotted out over 5 weeks, which is the amount of weeks my scraper was running and hopefully the amount of real life weeks that S52 took place over.

Now that explanations are out of the way, it is time for some fun graphs! First, we'll take a look at all the teams together so you can see how things all stack up.
[Image: IWRuUqf.png]

As you can see, most of the teams start out with an average TPE of around 275 or more. Newfoundland, Anchorage, Colorado, and Anaheim all start out with around a 275 TPE average. Significantly above that, we see Kelwona, Carolina, and St Louis starting with around a 300 TPE team average. And then, way down at the bottom, we have Detroit and Halifax with around a 225 TPE average. Just from looking at this first graph, we can make some guesses as to which teams have more experienced players and which teams are currently, let's say, "rebuilding."

Why don't we take a closer look at some of these graphs. We'll do it in alphabetical order for ease of navigation.

[Image: UtJxrw2.png]

Here we have Anaheim's Weekly Average TPE. This is one of the more straight forward graphs of the group. You can see that everyone on the team was gradually earning more TPE as the weeks went on and only making the weekly average higher. Setting a very great example, good job Anaheim!

[Image: INvJnZh.png]

Next we have Anchorage. The Armada started just below what Anaheim had and seemed to follow a very similar trend line. Thus, they finish just below Anaheim. This is another fairly straight forward graph. Continuous upward trend showing that the players continued growing throughout the season. Once again, a great example for other teams to follow. Way to go, Anchorage!

[Image: KSb5pim.png]

Ahhh, Carlolina Kraken. Our first "different" looking graph. So you'll notice how this one dips below 300 in week 3 and then starts to grow again. I can explain what has happened here, though you may have probably guess already. Between when I scraped for weeks 2 and 3, the Carolina Kraken added at least one new player. Rinako Yukikami joined the team and subsequently caused the the team average to go down. (I might also add that Mr. Yukikami's player page was so differently formatted from everyone else's that I had to write an additional function just to handle this one player.) You can see, though, that the team quickly seems to recover and continues growing throughout the remainder of the season. Also, Carolina is one of the few teams that started with over 300 TPE. This means that a large part of their team already had higher TPE and so the weekly average was growing at a slower rate than a tea filled with mostly rookies.

[Image: CIWT67n.png]

Colorado. It looks rough, but there is a very good explanation as to why the average TPE drops so much at the end. Colorado added three new players before the playoffs. These new players had relativity low TPE and brought the average down very far.

[Image: XnOZdeF.png]

Next up we have my own team, the Detroit Falcons. Detroit started with the lowest team average TPE in the league at just over 222 TPE. This is a very good indicator that the team is very young and inexperienced. The team continued to grow over the weeks, but despite their best efforts, they still finished the season with the second lowest average TPE, overtaking Halifax after week 2. An interesting this to note here, though, is that Detroit was able to finish 1st in the league during the regular season despite having an average team TPE almost 50 below the 2nd place team, Carolina. Perhaps their GM, @notorioustig, really knows what he's doing out there? Seems like a GM of the year sort of season to me....

[Image: NCJOLIY.png]

Halifax, like Detroit, started with a pretty low average TPE. Again, you can tell this is a very young team with a lot of growing to do. And grow they did! Halifax has a pretty steady growth rate, but not quite as quick as Detroit's. Halifax would finish with the lowest team average TPE in the league as well as in the regular season standings.

[Image: ZE5db0w.png]

Kelowna started with an very nice team average TPE of 300.00. They were amazingly able to grow to 311 by the third week. This is impressive because as you start getting closer to the 425 TPE cap, it is harder to grow at faster rates like the younger teams. Like several other teams, Kelowna took a hit when they added some new players, but this was no problem for them, as they were able to bring the average back up to 311 by the start of the playoffs. The Knights finished the season with the highest average team TPE, but finished at a quite low 7th in the regular season standings.

[Image: JfzbXaS.png]

Looking only at this graph, Newfoundland seems pretty average. They started out with a pretty average team average TPE and finished pretty average as well. Newfoundland finished 8th in the standings, which really isn't that bad consider how how they were nearly 30 TPE below the top contending teams' average TPE. That is really all I have to say about these guys. Maybe we will see some more growth from them during the off season.

[Image: 7iomFFW.png]

It's Scarecrow time. St. Louis was also one of the top teams in the league as far as team average TPE goes. Starting at just below 300 TPE and climbing all the way up to 310 TPE looks really good on paper. However, St. Louis only managed to finish 6th in the standings for the regular season. This is really interesting, as they never took any major TPE hits throughout the season like some other teams did, and they had a much higher TPE than some of the teams that finished ahead of them. I'm sure the success of the very low TPE Falcons is frustrating the the Scarecrows, especially after Detroit walloped St. Louis in the first annual SMJHL Winter Classic.

[Image: nxbLteZ.png]

Finally we come to Vancouver. First of all, I really like Vancouver's colors. They stand out well amid the mix or dark, red, or oranges that fill the SMJHL. Second of all, wow. Look at that growth between weeks 1 and 3. Bringing up the average TPE for your team by 20 in just three weeks is incredible. After that it seems to stagnate. I am not sure what happened there. It doesn't look like any new players were added to their roster, so perhaps players just went inactive. They still managed to grow, but not by much. Still, they finished 5th in the league for regular season and that is not easy feat while having a 30 less TPE average than the top teams.

In conclusion, these graphs can offer a lot of insight, but not into what areas you might think. You might think that these average TPE graphs could be used as good indicators as to how well a team is doing or going to do, but as we have seen, some of the highest average TPE teams finished very low in the standings and the number one spot at the end of regular season play went to the team with the lowest initial average TPE and the second lowest average TPE at the end of the season. This means that in this regard, the graphs are absolutely pointless. I believe that the best use for these graphs is actually analyzing player and team management. As for players, TPE doesn't mean much if it is allocated inefficiently. Teams like Detroit, whose GM is very hands on with guiding player builds, prove that allocating TPE correctly is very important to succeeding. In regards to team management, having a team of well built players doesn't necessarily mean they are going to work well together. The manager of the team has to do a good job of finding good player combinations and continue to improve them throughout the season. It isn't always going to be best to have all of your top guys on the top line. Again, I'll use @notorioustig as an example. There were times during the season that he had very low TPE players on the first line because he discovered that they work well with certain other players. I think that these graphs in conjunction with the results of the season show just how important it is to be hands on with your managerial duties as GM.

In short; I have the evidence laid out in front of you. Tig deserves GM of the Year.

Thank you for coming to my talk.

***notorioustig*** · 02-29-2020, 12:19 AM

Your flattery will get you nowhere

....except 1st line/PP

Nhamlet · 02-29-2020, 12:21 AM

All I see is that Simon spits at your TPE.

Mediocre_Fred · 02-29-2020, 12:24 AM

02-29-2020, 12:21 AMNhamlet Wrote: All I see is that Simon spits at your TPE.

Yes, that is my alternate analysis of this whole thing.

DeletedAtUserRequest · 02-29-2020, 12:42 AM

|

|

|

|

Mediocre_Fred · 02-29-2020, 12:48 AM

02-29-2020, 12:42 AMMike Izzy Wrote:

Thanks, I appreciate it!

***boom*** · 02-29-2020, 01:09 AM

This is awesome! It’s cool to see which teams were more active than others. Also curious to see Detroit finishing in first place with one of the youngest and lowest-rated rosters in the league.