AI Expert & Kaiju CEO Talk Stock Analysis & How AI Can Help You
This episode is all about information: where to get it, how to use it, and how AI can help make that whole process a lot easier for you
Aaron and Ryan cover key sources for reliable information - from free to paid - software tools that aggregate information, analysis tools, and where AI fits in all this as the fastest, cheapest way to cut through informational noise to reliably get to what you want to know.
If you always feel like you’re the “last to know”, then this episode is for you.
Aaron and Ryan cover key sources for reliable information - from free to paid - software tools that aggregate information, analysis tools, and where AI fits in all this as the fastest, cheapest way to cut through informational noise to reliably get to what you want to know.
If you always feel like you’re the “last to know”, then this episode is for you.
Category
🗞
NewsTranscript
00:00 [MUSIC]
00:06 >> All righty, Ryan. Welcome back to a Benzinga Live.
00:09 Hope you had a good New Year,
00:10 good holiday break and everything,
00:12 and it's good to see your face back on here again.
00:15 >> Yeah, it's nice to be back.
00:17 I think everyone appreciated the break from my weekly ramblings,
00:21 but hopefully we can add a couple more episodes,
00:24 a little bit of value here for everybody,
00:27 and it's always good to catch up with you, Aaron.
00:29 >> Yeah. Let's discuss real quick, Ryan,
00:34 where we left off in this educational series,
00:37 where we're at now,
00:38 and where are we headed?
00:40 >> Well, that's a lot to cover.
00:43 That's like six episodes.
00:45 I think we started with-
00:46 >> Give me the spark notes.
00:47 >> Yeah, we started with broker selection,
00:51 what type of account to open.
00:53 We talked about tools,
00:54 we talked about technical analysis,
00:56 we talked about risk management,
00:58 we talked about entering and exiting trades,
01:01 tools to do that.
01:03 Now today, I think we're all the way up to news,
01:07 and news with respect to stock analysis,
01:10 and where do you find stocks to trade,
01:13 and what news is relevant,
01:15 and what news is noise,
01:16 and all that good stuff.
01:18 >> Hundred percent. We were talking about,
01:21 when I started this show,
01:22 talking about these crazy headlines,
01:25 and it's important with all the headlines to know,
01:27 what's just noise,
01:29 what are the things you really should be paying attention to?
01:32 Let's get right into it, Ryan.
01:34 Let's talk information and analysis.
01:37 How do you go about actually
01:39 selecting the stocks you're interested in,
01:42 and what information are you looking for?
01:44 >> I think it depends on what investor or trader,
01:48 the person is who's doing the analysis.
01:51 If it's systematic, or quantitative, or technical,
01:55 then you're probably less concerned with headline news,
02:01 corporate actions, corporate news, etc.
02:03 You're looking for specific patterns that you've
02:06 identified previously as being profitable.
02:10 There's no yes or no,
02:13 or maybe with systematic trading,
02:16 it's like a series of criteria that gives you
02:18 a score at the end of the day.
02:19 If the score is above X, trade.
02:21 If it's below X,
02:23 then you don't trade and you just move on to the next stock.
02:26 You don't look for reasons to support it or not support it.
02:30 The most mathematical, that quantitative.
02:33 Technical trading, I think there's sometimes a misconception
02:38 that technical trading is just technical,
02:39 that nobody cares about earnings or news or whatever,
02:43 which isn't exactly true.
02:44 You're using technical signals and patterns to alert you that
02:48 there might be something advantageous or
02:50 disadvantageous about to happen,
02:52 and then you're going to go deeper,
02:54 depending on what type of trader you are.
02:55 You can be like a technical position trader.
02:58 So three, six months out,
03:00 probably that trader, after doing solid technical analysis,
03:04 is doing a deep dive into some more news, etc.
03:09 Whereas like a momentum or swing trader is looking for news
03:15 only as it relates to confirmation or a potential roadblock
03:19 that could lead to a whipsaw reversal, and that's it.
03:21 They're there for three to eight days, let's go,
03:25 and just moving on.
03:26 So I know everyone hates the "it depends" answer,
03:30 but yeah, it depends, which wouldn't surprise anyone
03:33 in this chat or watching this show.
03:36 Yeah, and I think that's a great point because a lot of times
03:39 people, and you'll see this from the gurus on Twitter
03:43 and other places, they try to act like there's one set of rule
03:48 or one thing that works for all traders.
03:50 But let's face it, we're going to have different goals
03:52 and we're going to have different capabilities.
03:54 Some people might be able to look at their screen
03:58 all throughout the day.
03:59 Some people might be going to work and they got to
04:01 maybe do a couple swing trades a month
04:04 and they're not able to day trade and all this stuff.
04:06 So you really need to figure out what's your style,
04:08 what works best for you.
04:10 And then Ryan, what tools are out there?
04:13 I mean, we hear a lot about the Bloomberg Terminal,
04:16 of course, is it just Bloomberg or Bust?
04:18 Are there any free sources?
04:20 Instead of having to pay 25 grand for a seat or whatever
04:23 on the terminal and how reliable are they?
04:25 - A lot to unpack in that one lengthy,
04:30 endless question there, Aaron.
04:31 But is it just Bloomberg or Bust?
04:33 No, I mean, Bloomberg is not the gold standard
04:37 of information, at least that's not my experience.
04:40 As I've said before on the show, yeah, fine.
04:42 We all use Bloomberg here, but we also use Benzinga.
04:45 Honestly, I am not just saying that 'cause I'm on your show.
04:49 You're looking for confirmation of news that leaks out.
04:54 And if it just appears in one platform or one area,
05:00 now if it's Benzinga, if it's Bloomberg,
05:02 you can have relative certainty that somebody has vetted
05:05 that at the news desk before it's been pumped
05:07 into the news feed.
05:09 And quick action can result in a little bit of extra profit
05:13 or the only profit on the day, that's totally possible.
05:17 But you don't need a paid source.
05:20 Is there free?
05:21 Yeah, sure.
05:22 But you're looking at the credibility of the source
05:25 and the identification of the source, right?
05:29 Like if it's an anonymous person in a Reddit forum,
05:33 if it's an anonymous Twitter account,
05:38 I would be heavily skeptical,
05:40 unless you know who that person is.
05:42 Like OpenOutCrier is a well-known options trader
05:47 that uses that handle on Twitter.
05:52 I'm dating myself there, I guess it's called X now.
05:55 People know who that is that are options traders.
06:01 Anyone who doesn't know who that is
06:03 wouldn't know that person,
06:05 but they're an industry professional.
06:09 If you can't identify the source,
06:11 then the jeopardy that person is under
06:15 in terms of market manipulation is very, very low.
06:18 So whether it's AI, whether it's a bot,
06:22 whether it's a malicious actor,
06:24 it's hard to determine, tell, and vet the source.
06:27 So I'd be heavily skeptical.
06:28 Like a company CEO or a public figure,
06:32 not that they don't make mistakes,
06:34 but you make a statement about company fundamentals,
06:38 you make a statement about confidence
06:41 in a stock's directionality,
06:43 and you are found to have manipulated the market
06:48 like you are in serious, serious trouble.
06:50 So news releases, even free sources,
06:52 Yahoo Finance or something like that,
06:55 or LowPay, which would be any sort of mainstream publication,
06:59 Financial Times, Wall Street Journal, both have alerts,
07:04 both have an aggregated newsfeed, PR Newswire,
07:08 AP, whatever it is,
07:10 before it hits these feeds, that news is getting verified.
07:16 It's not to say that it's not wrong.
07:19 It's that if it is wrong or worse misleading,
07:22 there's significant jeopardy behind it.
07:24 So it doesn't have to be paid.
07:27 You just have to do more heavy lifting yourself,
07:30 running down the verification of this information
07:34 than you would if you subscribe to an aggregator.
07:37 Money.net, let's do three, Money.net, Benzinga, Bloomberg.
07:41 Probably low to high in that order.
07:44 - Yeah, and I appreciate, I mean, Benzinga Pro,
07:49 I think for a lot of traders,
07:51 I think Bloomberg is that coveted,
07:54 what people hear about,
07:55 what they might see in the Wall Street movies or whatever,
07:58 but for most retail investors, which I am,
08:03 Benzinga Pro really does fit a lot of your needs.
08:05 I'll just pull it up real quick,
08:06 'cause we're talking about it.
08:07 So say you're looking at Nvidia stock here,
08:10 you have the overview.
08:10 If you wanna go into a more detailed view of the chart,
08:13 you can go through here
08:15 and go to all the different timeframes,
08:16 daily, weekly, monthly, hours, minutes.
08:19 The newsfeed, which is what a lot of people
08:21 use Benzinga Pro for, is actually super quick.
08:25 And like you said, Ryan, sometimes,
08:28 not all the time, not most of the time,
08:29 but sometimes you'll actually see something come through
08:31 on Benzinga's wire before Bloomberg.
08:35 And then of course, too,
08:36 I usually just hang out on this overview page
08:38 a lot of the times,
08:39 'cause you can see the chart
08:42 and all this fundamental data, market cap, PE,
08:45 all the good stuff here.
08:46 So again, for most traders,
08:48 and it depends on what you need,
08:50 but for most traders, I think Benzinga Pro does fit
08:54 those needs pretty well.
08:57 - Yeah, I mean, to your earlier point,
09:00 we run both feeds side by side.
09:03 I would say 98% of the time,
09:07 news shows up in both feeds simultaneously.
09:10 2% of the time, sometimes Bloomberg has it first,
09:13 but sometimes Benzinga does have it first.
09:15 I mean, if Benzinga didn't have it first,
09:17 we wouldn't run it, honestly.
09:18 Already being a heavy Bloomberg user,
09:21 we just stick to that.
09:22 There's a reason that we use Benzinga side by side.
09:26 Bloomberg doesn't always have everything.
09:29 And it's important to remember
09:32 that it's skewed towards the institutional investor,
09:37 fund manager, et cetera.
09:39 So there's millions of headlines a day
09:42 from all over the world.
09:43 Which ones are trickling to the top,
09:45 which ones are sort of marked
09:47 as being market moving on Bloomberg
09:50 are not the same things as what shows up on Benzinga.
09:53 You'll be able to find both, right?
09:55 The stories will hit both,
09:57 but that sort of top 100 aggregated news feed
10:00 is very, very different on Bloomberg
10:03 than it is on Benzinga.
10:04 Benzinga attempts to be skewed more towards
10:06 anything huge and global, obviously,
10:10 but stock specific, right?
10:12 There are a lot of short-term traders
10:16 who use Benzinga Pro
10:17 and they're looking for actionable signals,
10:20 news, et cetera, in the moment.
10:22 On Bloomberg, it's really more around
10:26 global market participants, global macro, et cetera,
10:30 like where interest rates shifting in Germany,
10:33 like this in a huge and diversified portfolio
10:36 is more important than
10:38 Nvidia's run this morning, for example.
10:44 I mean, that will show up in the same feed,
10:47 but that's one stock out of 972 positions
10:52 in just one of the portfolios that we run.
10:56 So yeah, it's important we pay attention to it,
10:59 but that's not as important as some of this,
11:02 some of these news items which aren't important
11:05 if you're a discrete stock trader.
11:06 So that's where, as I've said before,
11:09 I hope people don't think that they're missing out
11:11 by not having a Bloomberg.
11:12 It's really, it wouldn't make you a more profitable
11:15 or a better trader or investor at the average level,
11:19 for sure, you don't need it.
11:20 It's not designed for that.
11:22 - Well, let's, I mean, while we're talking about Bloomberg,
11:27 I do wanna hear a little bit more about it.
11:29 I'm not on the terminal, you are.
11:31 Let's give us some of like the inside ball
11:34 and what it's like and what we're looking at here.
11:36 I'll go ahead and pull up a slide of one of your homepages.
11:40 - Right, so for daily news,
11:42 like this is my daily view in term.
11:44 I have three 30 inch layouts that I'm looking at
11:48 and they're 6K layouts.
11:49 So apologies if the font is super small,
11:53 it's not as small in front of me.
11:56 And this would be my, as I'm doing something else,
12:01 email, Zoom meetings, talking to you here, it's right there.
12:06 This is what's up generally.
12:10 And so this is giving me kind of a global picture
12:14 of what's going in areas that tell me
12:18 where I need to better focus.
12:20 So it's important to remember that like the visual cues
12:24 of just like line charts are useless
12:26 for any kind of technical analysis.
12:28 Like all this stuff on the whole right-hand side
12:31 is pointless for any kind of actual analysis.
12:35 What's that useful for?
12:37 When everything drops in a geographic region,
12:41 Europe, I've got all of the major European indexes,
12:44 I've got the Asian index, domestic indexes.
12:47 When suddenly things move in harmony,
12:51 it's a visual cue that my attention needs to shift
12:54 to something, something's happened
12:57 to make everything move in unison.
13:00 So then my eyes would go from there.
13:02 I obviously have flow underneath.
13:04 I use index ETFs at the top there for domestic markets
13:09 just because I can better see flow
13:14 and that helps me with some of my analysis.
13:18 But that sort of unification of behavior
13:21 tells me I need to look to the left-hand side
13:23 of that screen, which is where I have like
13:25 aggregated news feeds and I can see which industries,
13:29 like you have sell up and sell down industries there
13:31 in that sort of center column.
13:33 And that's gonna tell me like,
13:35 as suddenly if that lights up, hot green, hot red,
13:39 heat map style, the two together,
13:43 I could be talking to someone, having a meeting,
13:46 half paying attention from, I don't know,
13:49 two feet away from that screen.
13:51 And I know that I need to look at something
13:54 if that dramatically changes.
13:56 Honestly, the Sky News or Bloomberg TV
14:00 that's in the upper left-hand corner,
14:02 that's like on, that got programmed in by our rep
14:06 and it's on mute basically all the time.
14:09 I will turn it on if there's like a breaking news item.
14:14 Biden walks out and he's about to make a live speech
14:18 on XYZ and it's fundamentally important.
14:20 Or when the Fed speaks, obviously I will turn
14:23 that on right there.
14:25 And then the map actually is a cool function
14:27 that allows you to select a tracking metric
14:30 and the map will highlight activity.
14:33 And it's everything from shipping distribution,
14:36 it's kind of on the left there underneath the TV, that guy.
14:39 Yeah, I mean, you can do forest fires,
14:41 you can do earthquakes, you can do whatever it is
14:45 and it will show you globally distributed,
14:48 like where concentrations are and something.
14:50 So that's, Bloomberg is not the only thing that does this,
14:53 but it's useful in that there's my,
14:56 I'm not really, I'm doing something else.
14:58 I need to be present, I need to kind of have my eye
15:01 on the portfolio, but I don't need to look at every tick
15:03 of every stock that we hold or trade.
15:05 So I'm looking at this, that's that.
15:07 And it's different from a screen that I might use
15:10 for news on a specific security.
15:12 So if something, which is the next screen,
15:15 and if something pops up in that aggregated news feed
15:19 that catches my attention,
15:21 or I get a message from our traders,
15:24 then yeah, it'll pop up in one of those two windows there,
15:28 or there's something that goes hot red or hot green
15:32 that I can click on, or the trade desk will ping me
15:37 on Messenger and say, hey, can you have a quick look at XYZ?
15:44 Then I pull up the next screen,
15:46 which is for the discrete.
15:50 So yeah, it's this one.
15:51 And we've talked before about the middle part of the screen.
15:55 I'm not gonna deal with that on this episode
15:59 because that's just technical analysis,
16:01 but it's the right side of that screen
16:04 kind of tells me at that underlying level
16:08 what I might be interested in paying more attention to.
16:11 So there's comps, there's a comp panel
16:13 that shows me the closest beta aligned comps
16:18 and how they're doing.
16:20 There's news specific to just that stock.
16:23 There's specific corporate action activity.
16:26 There's earnings activity, there are projections.
16:29 So it's like the global version,
16:33 but immediately targeted to just that security
16:37 and the ones that are really close to it.
16:39 I mean, at the bottom, there's,
16:41 you know, there's on the left as well,
16:43 there's even like flow identification
16:45 and trying to figure out where the axis on that day,
16:50 which is the heaviest exchange
16:52 for printing that stock on that day.
16:54 So I think Benzinga Pro
16:57 and a number of other retail accessible products
17:02 have these types of panels that you can consider.
17:05 So, you know, I do sort of the stock analysis
17:09 or a company analysis all on the one page.
17:11 There are tabs there.
17:12 You can see I can get through to the tear sheets
17:15 and company fundamentals as well on the right hand side.
17:18 And if that was trading,
17:20 there'd be order flow on the left hand side.
17:22 So it's like my command console,
17:24 it's everything that's salient visually immediately
17:29 with that particular underlying
17:32 and then the layers underneath that.
17:33 So, huh, I wonder what their revenue was quarter on quarter
17:38 over the last year.
17:39 That's a tab that I can drill down in.
17:40 I don't care about that right off the top.
17:43 I care because somebody sent me here
17:46 with usually a where do you see this staying
17:49 or what the hell's going on with this?
17:51 And I need to identify very quickly
17:53 what I think is happening there.
17:55 So that's this screen.
17:58 But again, you know, these are mostly tools
18:03 that are geared towards a user that's watching.
18:07 I mean, we look after four funds
18:11 and a number of portfolios in the funds
18:14 and different types of strategies.
18:17 So everything from, you know,
18:19 highly manual expert curated volatility arbitrage
18:24 through all of our AI curated and directed strategies.
18:28 And it's a different type of analysis for each.
18:33 You know, if you're laser focused on your portfolio,
18:37 you've got like 10 to 30 holdings,
18:41 you don't need half of this stuff.
18:43 I mean, it's helpful to have obviously
18:46 some visual representation of your concentration
18:49 so that, you know, if your portfolio is blowing up,
18:53 you can immediately see that that is tied to an overweight
18:57 in a specific area.
18:58 So it's not you, you didn't pick poorly.
19:01 This isn't like some horrible global mistake you made
19:04 other than being over concentrated in a sector
19:08 that's getting his face kicked in at the moment, right?
19:10 So that's helpful for you to manage your reaction.
19:14 And that's, I think what's important to keep in mind
19:17 when reviewing these type of tools.
19:18 - Yeah, and you mentioned earlier that it's not like,
19:22 you know, retail investors are SOL
19:26 if they don't have access to Bloomberg.
19:28 I mean, what do you think are like, you know,
19:31 we talked about Benzinga Pro obviously,
19:33 but what do you think are good other alternatives
19:36 to, you know, go in places you can get
19:38 some of this information,
19:40 and it might not be all available
19:42 or as quick as the Bloomberg,
19:44 but kind of, you know, like a good alternative
19:47 than if you don't have access to a Bloomberg seat.
19:50 - I think you've got sort of three key tools
19:54 that you could use.
19:56 I would for sure advocate for Benzinga
19:59 as a consolidated source.
20:00 Okay, I like it.
20:01 I think that's very close at a retail level
20:05 to what you could expect in a portion of a Bloomberg terminal
20:09 that would be meaningful to a retail trader or investor.
20:13 Okay, so you have that.
20:14 For deeper technical analysis,
20:18 I would use something like TC2000,
20:20 which we've showcased on this show before.
20:24 It also has a news aggregator,
20:26 but it's not as good as Bloomberg.
20:28 And, you know, Bloomberg's news aggregator is fantastic,
20:31 and the charting and technical isn't quite as good as TC2000.
20:35 So I'd like have them up next to each other.
20:38 But I think honestly,
20:40 I never advocate for trading off of trader chat room chatter
20:45 because it's just dangerous.
20:49 You can be in a circumstance of the blind leading the blind,
20:52 but if you are just starting out in trading,
20:56 who are you talking to?
20:58 I mean, if like all your friends don't trade all day long,
21:01 and if you are sitting at home trying to do this,
21:05 and you have nobody to turn to,
21:08 your cousin, uncle, father, best friend
21:12 is not an institutional trader,
21:15 does not work for a buyer sell side, market participant.
21:18 Like who are you bouncing ideas off of?
21:20 Who are you asking questions to?
21:22 It can be a pretty lonely environment.
21:25 I mean, like the CTO of Kaiju and I met in,
21:29 okay, so it was a little more of a pro trader chat room,
21:32 but it was still a chat room
21:34 because we were both trading remotely,
21:36 you know, him prop trading and me institutionally trading.
21:40 There's like nobody sitting next to me.
21:42 You know, you're craving this old school Wall Street thing
21:44 that died off years before, you know, people realize it did.
21:48 Like those old school trade desk,
21:49 trade floors are largely gone outside of institutions.
21:54 Remote work came in and was profitable and popular
21:57 long before COVID.
21:58 So you had these like satellite traders
22:01 dotted all over the place, especially APAC.
22:03 Like a lot of the guys who work Hong Kong, Singapore
22:06 were remote for the big banks
22:08 that just didn't have a regional office.
22:10 So who are you talking to all day long?
22:12 You know, you're going to a chat room to say,
22:14 hey, I've seen this thing,
22:17 or can someone tell me what they think about?
22:19 And eventually like you see people in the chat here
22:21 talking to each other, right?
22:22 There are a couple of these guys, gals
22:25 who clearly know each other from this chat.
22:28 Hey, how are you doing?
22:29 No, how are you doing?
22:30 How was your Christmas?
22:31 Yeah, good.
22:31 This looks interesting.
22:32 Yeah, I checked this out.
22:33 That's like, you need some cohort that you somewhat trust
22:38 to bounce things off of in the beginning.
22:42 So, I mean, I always come down hard on retail chat rooms,
22:46 but you know, they serve a purpose.
22:48 You know, I'm going long XYZ 300 at whatever,
22:53 don't do that, right?
22:55 Don't follow, that's not a good use, right?
22:57 You know, you follow it, you get burned,
22:59 you're pissed at that guy.
23:00 Why'd you listen to him in the first place?
23:02 But just kicking around ideas or saying,
23:05 this is the third time you've mentioned Bollinger Bands
23:07 and I don't know what those are.
23:09 And then maybe the person kicks you a link
23:11 to some educational resource.
23:13 So you become a better informed trader.
23:15 Like there's a use for that.
23:16 So that would be my three panels,
23:19 like a technical tool for evaluation.
23:22 I'd have a news aggregator
23:25 and I would have some communication
23:27 with the outside retail trading world.
23:30 If you want, and obviously you have
23:32 your broker platform there,
23:33 but you know, if you're using all of these other things,
23:36 your broker platform is gonna be like
23:38 an order ticket entry window, right?
23:39 It's just gonna be that part.
23:41 The rest of it will be like minimized.
23:43 So that would be my take.
23:45 If I didn't have a Bloomberg, I had to do it from scratch.
23:49 - Well, there you have it.
23:50 I mean, we've talked a lot about different sources,
23:54 where to get information, what information to look at,
23:57 how to help you consider which stocks to invest in.
24:01 And Ryan, you're the AI guy.
24:03 How does AI fit into all of this?
24:06 - I think, so AI assisted tools for retail traders
24:10 is for sure something that I know a number of outlets
24:14 are building actively right now.
24:16 The ones that have already come out aren't really AI.
24:19 Those are just scams.
24:20 So you can forget all of that.
24:22 And while, you know, generative AI,
24:25 LLMs are great for pulling in
24:28 enormous amounts of information,
24:31 they're terrible at math.
24:32 So you wouldn't wanna ask ChatGPT,
24:34 hey, you know, ChatGPT, what are good stocks to trade?
24:38 Please don't do that.
24:39 You know, how would I build an effective
24:42 and profitable portfolio?
24:43 You know, don't do that.
24:44 You know, we had fun with it
24:47 when the first beta became publicly accessible.
24:50 I asked, how do I replace my greedy vol arb traders?
24:55 And it was like a 10 point list.
24:58 And the first one was, first,
25:00 you will need to become an expert
25:01 in volatility arbitrage trading.
25:02 Then you can fire this desk
25:04 of lazy volatility arbitrage trader, blah, blah, blah, blah.
25:07 I did send that to them and they thought it was funny.
25:10 But the point being, like,
25:12 there's no way it's going to do that.
25:14 What you can do is try to scrape a lot of information
25:19 using a tool like that.
25:20 You can attempt, okay, so GPT,
25:23 I wanna know how many 13 Fs for Nvidia
25:28 were filed in the last quarter
25:31 where the top 10 market participants changed more than 20%.
25:36 Like, and it probably can return all of them.
25:39 You can give it a timeframe just to make sure.
25:41 You can say, in the last four years.
25:44 So you know how many 13 Fs you're going to get, right?
25:47 Four per year and however many years you go back,
25:51 that's how many you expect.
25:52 So if it gives you three,
25:54 you know that it hasn't collected all the information.
25:56 If it gives you what you think, you'll get the information.
25:59 You know, and tell me how many you found.
26:02 Very useful for something like that.
26:04 Very useful for you hear something
26:07 that you think sounds realistic,
26:09 sounds like it's from an untrusted source
26:12 in a bit of a sketchy, you know, venue,
26:15 but it sounds right.
26:17 You can task, you know, something like GPT
26:21 with finding any other references to this.
26:25 Can you find me all of the references
26:26 to the following piece of news and cite the sources?
26:30 And then you'll look, and if it's just fringe,
26:33 you know that this is some batshit conspiracy theory,
26:35 but maybe there was something buried in FT.
26:38 Maybe there's something buried in the Inquirer.
26:42 Maybe there's something buried in,
26:44 or sorry, Economist, why did I say Inquirer?
26:46 Don't listen to that.
26:48 Maybe there's something buried in a Bloomberg Newsfeed,
26:50 et cetera, et cetera, where you can verify this thing
26:53 that came from this sketchy source and actually act on it.
26:56 So it's very useful for that.
26:58 Unfortunately, on the side we work on,
26:59 the predictive AI side,
27:01 there's just nothing currently available.
27:04 And that's, you know,
27:06 something I hope that we can fix
27:07 in the next couple of years.
27:09 - Got it.
27:10 And then, you know, maybe just for people
27:13 that are joining us for the first time
27:15 on one of these educational series,
27:18 can you talk a little bit about how Kaiju is using AI
27:23 and what kind of, I guess, different types of things
27:28 you're feeding into the AI to help it decide
27:32 whether or not, hey, this is a dip we want to buy,
27:34 or no, this is not a dip we want to buy.
27:37 - You're right.
27:38 Okay, so sort of two main disciplines of AI
27:41 would be generative AI and predictive AI.
27:43 What's the big difference?
27:44 So there are no hard edges
27:46 to the landscape of generative AI, right?
27:48 I want this thing to create portraits of people.
27:52 So I feed it 600,000 portraits,
27:55 but I can't feed it all portraits
27:57 because all portraits isn't something
28:00 that is achievable.
28:02 It's just like I want it to speak English.
28:07 So I feed it, you know, 20 million samples
28:11 of the English language, but I can't feed it all samples.
28:15 And so as a result, you get hallucinations, right?
28:18 Mistakes that it believes is a correct answer
28:21 that it feeds back to.
28:22 With predictive AI, which is what we do,
28:25 it doesn't have the sex appeal on the surface
28:29 that generative AI does because it can't speak English.
28:32 There's no back and forth between you and the machine
28:35 that mimics a human interaction.
28:38 - Right.
28:39 - But the output we find vastly more powerful
28:42 because it uses known or collected at source data.
28:45 Price time and quantity for any equity
28:47 at the tick level is not up for discussion.
28:49 It is what it is.
28:51 It's printed to consolidated tape every day.
28:53 We've ingested 220 terabytes of this.
28:56 We ingest the stock market at the tick level.
28:58 So again, historically uploaded
29:00 and then collected at source.
29:02 So the patterns that it sees
29:05 can't be hallucinations in any way.
29:08 There are hard edges to that landscape.
29:10 And so it's predictive capabilities,
29:12 while not crystal ball like are pretty good.
29:15 So we've been doing this for a little over half a decade now
29:19 and are now, the reason I'm here is that, you know,
29:22 we stopped kind of flying quietly under the radar
29:25 like Renaissance Technologies
29:27 because nobody's talking about this part of the business.
29:30 And we wanted to try to, you know, educate a little bit,
29:34 which I think we'll get into with our next series.
29:38 This one was more about basic trading.
29:40 The next one is sort of more about advanced management
29:43 and using some of these tools.
29:46 So the criteria that it's looking for
29:49 are repeating patterns that it reliably links
29:52 to repeating behaviors, right?
29:55 Every time I see this or eight out of 10 times,
29:58 the following thing happens.
30:00 And of course, because there's a machine learning component
30:02 to it, as there are subtle micrometric changes
30:05 to the criteria and the behaviors,
30:08 that weighted list of endpoints
30:12 that it evaluates changes over time.
30:15 So maybe right now it's looking at 68 different factors
30:18 weighted from here to here.
30:21 But over time, it starts seeing that these top three or four
30:23 that it used to totally depend on
30:25 are less and less reliable
30:27 and less linked with positive outcomes.
30:29 These guys over here at the other end of the scale
30:31 have started to be more and more in the spotlight.
30:34 It can autonomously make those adjustments,
30:37 which is, you know, the rewarding and cool part
30:40 of working with predictive AI
30:42 is once you've programmed it to do this and it does it well,
30:45 I mean, certainly like six years ago,
30:47 we had instances of not doing it well.
30:50 And it was more horrifying to watch it drive off a cliff
30:52 in the boat full of money,
30:53 but thank God it stopped doing that like half a decade ago.
30:57 And now it's just profitable.
30:59 So we're hoping that like those will never get pushed out
31:04 to an end user in a retail environment,
31:06 because you're then creating like a trade system
31:09 that somebody else is following.
31:11 What I hope is that the predictive element
31:14 becomes available like GPT
31:16 for retail trader investor to query.
31:19 I've seen this pattern.
31:21 You know, I want to trade XYZ in this direction.
31:24 Here's my profit target.
31:26 Here's my stop loss.
31:28 What are my chances of success?
31:30 And then, you know,
31:31 2 billion calculations in one second later,
31:35 you get up not very good.
31:36 But if you did this, this, and this, and this,
31:39 it increases to that.
31:40 Like I would love to see that available,
31:43 those types of tools for, you know,
31:45 people like those who are watching this show.
31:49 - Yeah, and I'm excited again next week to hop into,
31:53 you know, a little bit more of that.
31:55 You know, I thought today we had some great lessons
31:58 just in terms of where to go for that information,
32:02 how to kind of sort through all this different noise.
32:05 It is, I mean,
32:06 I don't need to tell you guys out there in the audience.
32:08 It is very overwhelming, especially for newer traders.
32:11 When you look at a screen,
32:13 even like Benzinga, I mean, even like this from Benzinga Pro,
32:15 which now that I'm on it every day,
32:17 this looks like such a simple, easy layout,
32:19 but I can see how something like this
32:21 with so many different numbers, so many different things,
32:24 just seems like too much.
32:26 And so, you know, this was a very important episode
32:29 to learn how to kind of cut through that noise,
32:32 how to use this information
32:34 to take your trade to the next level.
32:35 'Cause without the information,
32:36 if you're just trading on your own,
32:38 off gut feeling or vibes, you're gonna get burned.
32:41 There's no way around it.
32:42 And you need, you know,
32:44 to be able to have a systemic trading system
32:49 with this information.
32:51 So Ryan, let's go ahead and pull up
32:53 our Kaiju Kicker for the week.
32:55 I'll start by reading the first one
32:59 and maybe you can add some extra color
33:00 or walk me through them.
33:01 But do use more than one news aggregator.
33:05 They usually carry all the same news, but not always.
33:09 - Yeah, so like we said before, you know,
33:11 with the Bloomberg versus Benzinga,
33:13 I mean, again, 98% of the time, it's the same,
33:17 2%, it's not.
33:19 That 2% repeatedly has been actionable for us.
33:23 So it's worth having two.
33:24 And it doesn't matter whether it's, again,
33:27 I don't care what aggregators you use.
33:29 Bloomberg could probably be replaced with something else
33:32 that is more meaningful on a retail level,
33:34 more effectively for way less.
33:36 So just like, don't just take what comes through the feed.
33:40 Benzinga is fine for primary news aggregator.
33:43 If you're gonna commit serious dollars to this,
33:46 check the source, which is number two.
33:48 - Yep, do check sources.
33:50 A rumor isn't always something
33:52 you should be trading off of, nor is a post in a form.
33:56 For example, I'll go, you had some color on this,
33:58 but you just yesterday on on stock ticker ONON
34:02 was trading higher on like speculation
34:05 that Tiger Woods was gonna sign a partnership with the,
34:07 it's like a sneaker company out of Switzerland
34:10 that Roger Federer has backed.
34:13 But there was rumors that Tiger
34:14 was gonna sign a deal with them after he left Nike.
34:17 And people were sharing it all over Twitter.
34:19 The stock was up like eight, 9% in the morning.
34:22 And then their CEO had to say,
34:25 "No, we're not signing a deal with Tiger."
34:27 And then the stock went back down.
34:28 So if you were just trading this news
34:30 based off these rumors on Twitter and other things,
34:33 then you could have gotten burned on that.
34:35 - Yeah, I mean, if you read any of the classic trading
34:40 treatises, like you'll come across this phrase,
34:43 "Buy the rumor, sell the news."
34:44 Like that's an old trader phrase,
34:48 but that's like long dead.
34:50 That's before social media.
34:52 That's where the rumor came from company insiders
34:56 and institutional market participants.
34:58 And the news came from a trusted news source.
35:01 Now the rumor is coming from who the hell knows where.
35:04 And then the news confirms or denies that rumor.
35:08 So it's a lot more volatile and not as safe to do that.
35:12 - 100%.
35:14 And then moving down our kicker, number three,
35:17 do not trade off tips shared on social media.
35:21 We kind of just touched on that.
35:22 Do not trade off tips shared in trading chat rooms.
35:27 And yeah, I mean, you only need to get probably burned
35:29 once or twice on stuff like this.
35:32 And then you realize, I mean, I can't tell you
35:35 how many times, Ryan, I've been in different chat rooms
35:37 or whatnot.
35:38 And someone's been like, "Oh, this biotech company
35:41 is about to get this approval after market today.
35:44 Buy today or you're gonna miss out on this move."
35:46 And then the thing doesn't happen.
35:47 And it feels like every time the more certain people are
35:52 on something happening or the more people get involved
35:55 in it, it's like, I know this isn't gonna end up
35:57 working out 'cause the market's not that easy.
35:59 You're not gonna have really these like get rich
36:02 quick opportunities.
36:03 Oh my God, there's this hot tip everyone buy.
36:07 And so that's definitely an important one just to remember
36:11 that even though some of these tips might look juicy
36:14 on social media and chat rooms, a lot of times
36:17 they don't end up panning out.
36:18 - Yeah, exactly.
36:20 And I'm sure like you probably have a ton of rules
36:23 for this chat that's running right now.
36:26 Some of the trading, like the actual trading chat rooms
36:29 that are accessible to retail traders will have rules
36:32 like you can't post an exit if you haven't previously
36:35 posted your entry to stop people from being like,
36:38 you just made a cool 100% return on XYZ.
36:41 It's like, you have to have previously in the chat said,
36:44 I'm in at here and out at here to even post it
36:47 or you get blocked, muted, banned, et cetera.
36:50 I'm positive if somebody started like pumping stocks
36:54 and wild claims of their trading in here,
36:58 they'd be gone pretty quickly.
36:59 So, bear in mind that not all chat rooms are created equally
37:03 I guess it's fair to say.
37:05 - Yep, always be careful where you get your information
37:08 including from chat GPT because I have had many times
37:12 where chat GPT has tried to tell me something
37:16 as if it is a fact and it is undeniably not a fact
37:21 and then I'll tell it, hey chat GPT, why'd you say this?
37:24 When that's not true and they'd say, oh, I'm sorry,
37:26 you're correct, Mr. Aaron.
37:28 So with that said, that's not even with financial stuff.
37:32 That's just with like, if I ever use chat GPT for anything,
37:35 I feel like it always makes at least one mistake.
37:37 So with that said--
37:38 - Have you ever had a company contradict itself
37:40 in the same response?
37:42 - Oh yeah, all the time, yeah.
37:43 - I can't remember, it was a while ago
37:45 but I think I asked it just for fun.
37:47 I said, hey chat GPT, is Overstock a good company?
37:52 And it like, the first part was like,
37:54 Overstock is a good company because blah, blah, blah, blah,
37:57 and then obviously it had found all of the Patrick Byrne
38:00 nightmare scenario crap.
38:02 It's like, Overstock is not a good company
38:04 because like that was its outcome.
38:06 It was like, yes, wait, no, totally, I'm broken.
38:11 And that was, that's where it ended off.
38:13 So yeah, please don't ask it what to trade, it doesn't know.
38:17 - Yeah, not a good source for trading tips.
38:20 Well, Ryan, we talked about a lot today.
38:25 If you guys are watching and you want to go back
38:29 and rewatch any of the parts, you will be able to do that
38:31 on this YouTube video.
38:33 And we kind of teased what we're talking about next week,
38:36 but do you wanna remind the folks
38:38 what we'll be getting into next Tuesday?
38:40 Same spot, same time, 11 a.m. Eastern.
38:43 - We were originally gonna talk about
38:45 sort of a walk around the market,
38:47 what market participants do, what direct market makers do,
38:50 what lead market makers do, et cetera, et cetera.
38:53 Then I thought, God, do we really wanna put people to sleep
38:56 at that level?
38:56 Is it completely meaningful
38:58 to the retail trader and investor?
39:00 So instead, I thought I would share like the seven things
39:05 that for sure I can correlate
39:09 to increased performance in my career.
39:12 That I think, and none of them require
39:14 any institutional connectivity whatsoever.
39:18 So I think that might be fun to close off by,
39:22 there are areas that you wouldn't normally think
39:24 they would be, but I can guarantee
39:27 that they will make you a more effective trader
39:29 and investor, no question.
39:31 - Well, there you go, Ryan.
39:32 Again, Ryan Pennell from Kaiju.
39:35 Where can folks go more
39:36 if they wanna learn more about Kaiju
39:38 or if they wanna invest in the dip ETF?
39:41 - Kaiju, the global companies are at kaiju.ai.
39:48 Our AI directed asset management is at KCM,
39:53 kaijucapitalmanagement.ai.
39:56 And dip is at dipetf.com or kaijuetfadvisors.com,
40:01 which is the advisor for the dip ETF.
40:05 - Well, there you go.
40:06 All right, Ryan, we'll chat next week.
40:08 Thanks again for hopping on.
40:09 It was good to be back after a few weeks off.
40:11 Good to see your face, hope all is well.
40:13 And again, next week, same place, same time,
40:16 11 a.m. Eastern.
40:17 We'll be back getting into it.
40:19 - Thanks, Aaron.
40:20 It's good to see you and I'll see everybody next week.