Full regression results

This is a technical post for people interested in details. The punchline is: even if you include a wide variety of possible predictors of success, a simple least-squares model looking only at likes is quite similar in performance to more complicated models using followers, tags, etc.

I don't think the information here is a very wide interest, so I'm just going to present results and link to my GitHub for those who want details. Feel free to ask questions in the comments if you would like more info.

Lasso Ridge OLS OLS-Likes only
1 0 0 2854.06965 N/A
Shares 0 11359.86 65.6032316 N/A
Comments 0 222.1719 -10.255222 N/A
Likes 4.800058 -422.25 4.2506603 5.79909414
Follower Count 0.071327 -15296 0.09265454 N/A
Following Count 0 -21.6523 0.67554451 N/A
Bio is empty 0 2395.795 2326.99936 N/A
Any Hashtags 0 20106.24 20100.5325 N/A
Any tagged users 0 814.3656 60.5159362 N/A
Is Commerce 0 0 0.00016675 N/A
Expected Views -0.1217 -11620 -0.1377142 N/A
Shares Comments -5.29E-06 -487936 0.00025447 N/A
Shares Likes -2.13E-05 -2.6E+07 -4.45E-05 N/A
Shares Follower Count 5.56E-06 -2.2E+08 2.02E-06 N/A
Shares Following Count 0.010245 -48044 0.01210208 N/A
Shares Bio is empty 0 -291.788 -259.11236 N/A
Shares Any Hashtags 0 -50.6875 -47.858801 N/A
Shares Any tagged users 0 -3.71094 1.38718181 N/A
Shares Is Commerce 0 0 1.87E-07 N/A
Shares Expected Views -1.68E-06 -1.9E+08 -3.38E-06 N/A
Comments Likes 1.09E-05 -2.7E+07 1.69E-05 N/A
Comments Follower Count -9.39E-06 -2.9E+08 -5.99E-06 N/A
Comments Following Count 0.008986 -69062 0.00768233 N/A
Comments Bio is empty 0 226.1685 226.214091 N/A
Comments Any Hashtags 0 -4.87109 -1.6125186 N/A
Comments Any tagged users 0 -2.25977 3.08488467 N/A
Comments Is Commerce 0 0 2.91E-08 N/A
Comments Expected Views 5.72E-06 -2.2E+08 1.68E-06 N/A
Likes Follower Count -9.75E-08 -1.4E+10 -1.05E-07 N/A
Likes Following Count 0.000318 -2870016 0.00031629 N/A
Likes Bio is empty 0.740149 8.395508 7.14095384 N/A
Likes Any Hashtags 1.820684 -213 2.37077498 N/A
Likes Any tagged users 0.165318 -428 0.27059424 N/A
Likes Is Commerce 0 0 1.55E-09 N/A
Likes Expected Views 2.07E-08 -1.1E+10 1.21E-07 N/A
Follower Count Following Count -1.23E-05 -3.7E+07 -1.54E-05 N/A
Follower Count Bio is empty 0.760261 14.51758 0.1814774 N/A
Follower Count Any Hashtags 0 -1776 -0.0198163 N/A
Follower Count Any tagged users -0.02367 -3800 -0.03748 N/A
Follower Count Is Commerce 0 0 1.30E-10 N/A
Follower Count Expected Views 3.89E-08 -1.3E+11 3.44E-08 N/A
Following Count Bio is empty 0 1.519531 0.71076586 N/A
Following Count Any Hashtags 0 -7.49829 -8.1253227 N/A
Following Count Any tagged users 0 9.777588 10.2881494 N/A
Following Count Is Commerce 0 0 6.60E-14 N/A
Following Count Expected Views -0.0001 -2.7E+07 -8.71E-05 N/A
Bio is empty Any Hashtags 0 -13448.7 -13558.681 N/A
Bio is empty Any tagged users 0 -4826.69 -4157.9899 N/A
Bio is empty Is Commerce 0 0 0 N/A
Bio is empty Expected Views 0 8.77417 8.39756356 N/A
Any Hashtags Any tagged users 0 -3487.8 -2919.8266 N/A
Any Hashtags Is Commerce 0 0 0 N/A
Any Hashtags Expected Views -0.03061 -988 -0.0065345 N/A
Any tagged users Is Commerce 0 0 0 N/A
Any tagged users Expected Views 0.012779 -2536 0.00638975 N/A
Is Commerce Expected Views 0 0 0 N/A
Index 0 -0.50781 -1.8726071 N/A
Intercept 4659.147 1.82E+23 2852.07352 15835.8962
Alpha 1E+08 0.001 N/A N/A
R^2 0.923156 0 0.92749566 0.87892063

Comments

Post a Comment

Popular posts from this blog

Skincare products recommended on TikTok

I recently complained that skincare products don't list their dosages. Several of you responded with companies which do list dosages. Here is a table comparing them. If you just want me to tell you what to do Here are the products I bought. They are basically the cheapest set of products which has closed the recommended dosages of the four best tested ingredients. The affiliate links give me a 1% commission. I of course appreciate you using them, but I also provide non-affiliate links, and the recommendation isn't based on affiliate fees. Product Amazon link (affiliate) Amazon link (non affiliate) When to apply Retinol The Ordinary Granactive Retinoid 2% in Squalane The Ordinary Granactive Retinoid 2% in Squalane Nightly Alpha-hydroxyl acids Good Molecules Overnight Exfoliating Treatment Good Molecules Overnight Exfoliating Treatment ~2x/week Vitamin C The Ordinary Ascorbic Acid 8% + Alpha Arbutin 2% The Ordinary Ascorbic Acid 8% + Alpha Arbutin 2% Mornings (may interact with r
Read more

How important is the first video you create on TikTok?

Image
Summary : There is a widespread belief that the first video one creates on TikTok 1) gets an automatic “boost” from the recommendation algorithm and 2) has extra importance in influencing future video success. I test these claims on a data set of 33,925 videos and find that both propositions are unsupported by the evidence. If you have been holding off on creating your first video because you want it to be perfect, you should just go ahead and make an imperfect one. Background I sometimes hear people argue that the first video one creates on TikTok is unusually important. For example : The first video that any user on Tik Tok posts gets an instant spike in impressions. As a new user, Tik Tok metaphorically “throws you a bone” and will in most cases show your video to more people right off the bat than it would a video by a repeat poster. This means that it’s vital to make your very first post on Tik Tok engaging. Unlike Youtube, for example, where you can feel free to po
Read more

Are likes, shares, or comments more important for TikTok video success?

Image
Summary TikTok’s recommendation algorithm is an iterative process which repeatedly displays videos to larger and larger test audiences, until engagement from the test audience falls below some threshold. It is unknown how “engagement” is measured, but it is usually speculated to be some combination of likes, comments, and shares. Various people have hypothesized the weights of these metrics (e.g. one share = 10 likes), but, to our knowledge, no one has ever measured it. We therefore attempted to measure this on a representative sample of ~20,000 videos. Our research concludes that these metrics are highly collinear and no predictive value is found from comment and share count. (I.e. you should just optimize for likes, and don’t need to optimize for shares or comments .) You will need about 1/5 of your viewers to like your video in order for it to be promoted . Background and Motivation More background on the structure of the TikTok algorithm can be found here . Veed cont
Read more