A Few More New Changes

Over the past couple of weeks since I announced some major changes after my upcoming vacation, I have thought of a few tweaks to things that I need to make to enhance the changes that I am making.

Here’s what I am doing in addition to the other changes:

  • I will aim for a minimum of 6 songs on the Level 10 Poll. With the change of only one song being promoted from Level 9 to Level 10 every week, I have figured out that we could eventually have a very small number of songs on Level 10 (especially with another tweak later on this list). In order to combat this, for any week where there is less than 6 songs on the Level 10 Poll, I will promote 3 songs from Level 9 to Level 10. I am okay going into a voting with 5 or fewer songs on Level 10 and I want to advertise on the survey when the number of songs going from Level 9 to Level 10 is different instead of increasing the number because of the results of the voting.
  • Songs on Level 10 have more strict requirements to stay in the voting. One thing I have noticed this year is that there are songs coasting along on the Level 10 Poll for 10 or more weeks that are not finishing last or not finishing high enough to really threaten the top songs. While I don’t want to force any song to have to finish a certain spot when its still new, I do think once a song has been in the voting for 20 weeks, it needs to do more than coast into the next voting. Once a song reaches week 21, it will have a similar restriction as the #1 song. Currently the #1 song on the previous chart has to finish in the Top 3 on Level 10 in order to remain in the voting. For songs that are not at #1 and are at least on their 21st week in the voting, they will have to finish 4th or higher to remain in the voting.
  • The number of points a song “gives up” on its “encore” week will no longer be strictly 10%. With me not needing a set number of points overall, I thought about how uniform songs were falling and how that is not how it works on the major charts. I originally decided upon 10% just to have more points available to spread out during the next voting, but since this is no longer the case, I can do something different. It dawned on me that I can use the results of the previous voting to determine how many points a song will lose. I have decided that the percentage of points a songs gives up will be the following:
    Songs reaching 30 weeks in the voting: I feel these songs have earned the right to not give up any of the points they have earned, so they will not lose points for their encore week chart.
    Level 10 Songs: any song that meets one the Level 10 requirements (finishing last, #1 song outside Top 3, and the previously mentioned rule), will lose 10%, rounded up to the next point (i.e., 10.4 points would be 11 points).
    Percentage: Songs that do not meet their percentage requirements that have points will now drop a minimum of 5 points and a maximum of 50% of what they had. What I am going to do is take the percentage they fell short and multiply that by 10. For example, if a song fell 3.5% short of what it needed, I would drop 35% of it points, rounded up to the next point. If a song missed it by more than 5%, then I will cap it at 50% just so they do not give up more points than they have left.
  • There will be a new “Most Accumulated Points” record. I originally decided that both the yearly chart and the record for Accumulated Points would both remain unaffected, but I have decided they need changed. Since there will consistently be about 1,700 points awarded each week, there is the potential for there to be a lot more points available than the current 10,000. Therefore, I will change the current record category to “Beginning of 2018 to 8/25-8/31/19) and create a new category “9/1-9/7/19 on.”
  • There will now be a tweak in point recording for the 2019 Yearly Chart. I would have rather waited until 2020 started to make these changes, but some of them help me save time so the sooner the better. I am going to create a “scale down” for the songs who peak on the chart even if they only received points in the new system once. I will determine that “scale down” at the end of the year as I will compare the data from the first 8 months to the data for the last 4 months and see what works best
  • Finally, I am making it a rule that if a song is in the Top 5 on Levels 1 through 9 it will not be retired. Let’s say there are 2 really, really popular songs on the Level 7 Poll one week and those 2 songs combine to get over 50% of the votes. This has a negative impact on the other songs on the voting to the points where it is possible for every other song not to reach the percentage they need. I am giving more points to 4th and 5th than the rest of the songs on that Level, so I am going to add this bonus so all promoted songs and a couple more are safe each week. If I don’t do this, there could be the chance that a Level could end up with a small number of songs. If there is a small number of songs, there becomes a poll where every song gets promoted every week with nothing I can do about it. Between this change and the one I made to Level 10, I am putting a “minimum” on the Levels to ensure the system will work.

Well, there are the rest of the changes that I could think of to help make the new, easier-to-understand and easier-to-maintain system work. I believe in letting everyone know the changes, but I also don’t want to make it so you have to know a bunch of stuff to understand the voting and the chart. I am going to create a page on the blog here that will outline what you need to know so you can understand the chart, the voting and how they both work.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s