It’s been more than a year since we posted our last list of people to follow on Twitter for financial news. Time for an update!

Without further ado, here is this year’s list (click on headers to re-sort):

Rank Screen Name Name Influence Score Relevance Score Order by Topic
1 pdacosta Pedro Nicolaci da Costa 6.78 7.39 149
2 TheStalwart Joe Weisenthal 10.0 2.48 122
3 ReformedBroker Downtown Josh Brown 8.87 2.99 87
4 ritholtz Barry Ritholtz 7.68 4.1 86
5 felixsalmon Felix Salmon 8.82 2.73 470
6 Noahpinion Noah Smith 5.12 6.07 310
7 Frances_Coppola (((Frances Coppola))) 3.64 7.48 49
8 hblodget Henry Blodget 6.43 4.59 252
9 crampell Catherine Rampell 5.26 5.34 338
10 sdonnan Shawn Donnan 3.23 6.94 366
11 moorehn Heidi N. Moore 3.96 6.06 401
12 paulkrugman Paul Krugman 6.91 2.44 358
13 M_C_Klein Matthew C. Klein 7.45 1.74 333
14 davidmwessel David Wessel 5.96 3.22 364
15 matt_levine Matt Levine 8.36 0.81 121
16 tracyalloway Tracy Alloway 8.45 0.62 123
17 IvanTheK Ivan the K™ 5.78 3.29 263
18 edwardnh Edward Harrison 4.01 4.75 146
19 Nouriel Nouriel Roubini 6.29 2.42 183
20 BCAppelbaum Binyamin Appelbaum 6.98 1.7 351
21 carlquintanilla Carl Quintanilla 5.19 3.33 261
22 acrossthecurve across the curve.com 0.92 7.53 131
23 MarkThoma Mark Thoma 5.09 3.27 317
24 AdamPosen Adam Posen 4.18 4.07 295
25 JustinWolfers Justin Wolfers 6.75 1.48 356
26 jennablan Jennifer Ablan 5.48 2.75 162
27 elerianm Mohamed A. El-Erian 5.85 2.15 151
28 tomkeene tom keene 7.09 0.87 148
29 CardiffGarcia Cardiff Garcia 6.57 1.31 332
30 jasonzweigwsj Jason Zweig 6.28 1.56 114
31 delong Brad DeLong 🖖🏻 5.2 2.62 345
32 ObsoleteDogma Matt O’Brien 6.6 1.23 355
33 greg_ip Greg Ip 6.67 0.94 107
34 RobinWigg Robin Wigglesworth 5.5 2.1 5
35 JacobWolinsky Jacob Wolinsky 2.29 5.31 238
36 TimOBrien Tim O’Brien 2.94 4.63 442
37 izakaminska Izabella Kaminska 7.34 0.1 178
38 tylercowen tylercowen 5.62 1.8 307
39 NateSilver538 Nate Silver 6.52 0.87 344
40 jbarro Josh Barro 4.7 2.65 339
41 jessefelder Jesse Felder 2.04 5.23 130
42 ezraklein Ezra Klein 6.0 1.26 340
43 EddyElfenbein Eddy Elfenbein 4.16 3.05 138
44 johnauthers John Authers 6.08 1.13 1
45 TabbFORUM TabbFORUM 0.2 7.0 42
46 AmyResnick Amy Resnick 2.32 4.75 394
47 activiststocks Activist Stocks 0.81 6.13 198
48 eisingerj Jesse Eisinger 4.77 2.14 435
49 TimHarford Tim Harford 4.37 2.47 14
50 Neil_Irwin Neil Irwin 6.1 0.71 350
51 Jesse_Livermore Jesse Livermore 4.69 1.97 89
52 carney John Carney 5.72 0.91 265
53 mims Christopher Mims 🎆 3.46 3.13 448
54 conorsen Conor Sen 3.96 2.61 142
55 lisaabramowicz1 Lisa Abramowicz 4.32 2.16 128
56 JohnCassidy John Cassidy 3.92 2.55 441
57 economistmeg Megan Greene 5.12 1.34 292
58 Austan_Goolsbee Austan Goolsbee 4.67 1.78 357
59 lopezlinette Linette Lopez 2.75 3.7 264
60 prchovanec Patrick Chovanec 2.2 4.22 279
61 mark_dow Dow 5.23 1.18 262
62 DLeonhardt David Leonhardt 5.72 0.65 354
63 niubi Bill Bishop 2.35 3.89 275
64 interfluidity Steve Randy Waldman 4.4 1.84 304
65 karaswisher Kara Swisher 3.55 2.68 453
66 tomgara Tom Gara 2.45 3.74 375
67 Convertbond Lawrence McDonald 5.48 0.64 135
68 AlephBlog David Merkel 2.13 3.96 129
69 SamRo Sam Ro 3.9 2.19 137
70 katie_martin_fx Katie Martin 4.51 1.54 23
71 JimPethokoukis James Pethokoukis 3.59 2.44 368
72 EpicureanDeal TED 3.05 2.97 411
73 TimDuy Tim Duy 3.18 2.83 144
74 abnormalreturns Tadas Viskanta 4.78 1.22 94
75 ianbremmer ian bremmer 4.95 0.92 284
76 rortybomb Mike Konczal 3.92 1.94 303
77 matthewstoller Matt Stoller 1.79 4.07 378
78 georgemagnus1 George Magnus 3.87 1.98 281
79 Alea_ JC Kommer 1.87 3.96 170
80 ModeledBehavior Adam Ozimek 3.55 2.24 325
81 PekingMike Mike Forsythe 傅才德 1.21 4.57 287
82 kadhimshubber kadhim (^ー^)ノ 2.16 3.59 27
83 morningmoneyben Ben White 3.89 1.83 428
84 NinjaEconomics Ninja Economics 2.17 3.54 312
85 BaldwinRE Richard Baldwin 2.45 3.25 297
86 valuewalk ValueWalk 3.49 2.2 197
87 LaurenLaCapra Lauren Tara LaCapra 2.17 3.52 404
88 ryanavent Ryan Avent 5.04 0.63 331
89 D_Blanchflower Danny Blanchflower 3.28 2.36 150
90 danprimack Dan Primack 3.97 1.67 475
91 cullenroche Cullen Roche 4.58 1.03 88
92 scottlincicome Scott Lincicome 1.11 4.46 367
93 MikeIsaac rat king 2.58 2.87 454
94 jmackin2 James Mackintosh 4.81 0.64 17
95 LorcanRK Lorcan Roche Kelly 4.93 0.51 67
96 alex in Providence alex 0.72 4.72 477
97 MattGoldstein26 Matthew Goldstein 2.65 2.79 431
98 AnnPettifor Ann Pettifor 2.52 2.88 50
99 modestproposal1 modest proposal 3.91 1.48 243
100 rodrikdani Dani Rodrik 4.26 1.12 318

A word about methodology:

1) Start with a few highly-followed accounts, e.g.: pdacosta, TheStalwart, ReformedBroker, ritholtz, felixsalmon

2) Determine who they follow! Traverse the Twitter graph using the API.

3) From these 5 people you can get a pretty great starting list:

  • About 14,700 users
  • But only 112 users who are followed by all 5, or 4 out of the 5

4) That would be a great place to start, but we can do a little better:

  • Cull the non-financial users, like darth or <insert political figure>
  • Iterate and create a new Twitter graph starting from the users remaining

In general, from a given list of users, get a better list by

1) First expanding it, by finding who these users follow

2) Then ranking and filtering the new list by

  • Influence: how many people in the list follow them, and recursively how influential they are (PageRank)
  • Relevance: how frequently they post financial content (financial sites, tickers, topics, and recursively, items that hit the StreetEYE frontpage)
  • Timeliness: how often they are first to post something that later gets popular

Iterate a few times, and you get a pretty good list of people to follow.

In the past I’ve generated a graph of the users, and this year I really went ham on it and created this magnificent beast:

If you click here you can explore the graph interactively:

  • Roll over and get detailed info on each FinTwit personality
  • Word cloud (roll over)
  • 3 most similar accounts, using topic analysis of what they post about, who they share same URLs as, who they follow and are followed by. P.S. I LOVE THIS FEATURE!
  • Each user’s most frequently shared domains, hashtag, tickers, other FinTwitterers they mention
  • Who they follow/are followed by (roll over ‘followed by’/’followed’)

I hope this helps everyone find great new FinTwit BFFs to follow.

It’s always a bit arbitrary, where to cut off people who aren’t relevant. Some people may find the influential tech or political accounts a bore, but I try to find a balance.

The biggest problem is churn. There are some people who are highly followed who don’t really post relevant stuff any more. There are people who are pretty relevant but it takes a very long time to break through and get influential. I could expand the panel, but the more you expand it the more the common denominator is… all Trump all the time.

Then, I guess I could use topic analysis to try to downvote Trump and politics… use noise cancellation to determine what is popular out in the broad population and penalize it … but it’s turtles all the way down the rabbit hole.

That’s it! If you’re looking for top blogs for your daily list or Feedly reader check out the July listicle of the most shared financial blogs.

Not to be too thirsty but if you like it don’t hesitate to share!

(Never say never again, but very probably never doing this graph again. Fun mad science, but disproportionately time-consuming. Twitter makes the API more restrictive every year, knucklehead sites block me, the graveyard is full of algorithmic news apps and news bots. C’est la vie!)