October 24, 2024 • 23 mins
In this episode, Jim Jockle dives into the transformative role of AI in trading. Joining him is Rance Masheck, CEO of iVest+, and Chris Mercer, COO and Head of Business Development for iVest+.
iVest+ uses an innovative AI-powered toolkit to reshape trading for their users with smarter insights and streamlined decision-making. From advanced screeners to personalized trading coaches, explore the AI strategies that are currently redefining the trading landscape.
There was so much information, we had to break it into a two-part episode. This is part one, stay tuned next week as we release part 2 of this insightful conversation!
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:07):
Welcome to Trading Tomorrow Navigating Trends in
Capital Markets the podcastwhere we deep dive into
technologies reshaping the worldof capital markets.
I'm your host, jim Jockle, aveteran of the finance industry
with a passion for thecomplexities of financial
technologies and market trends.
In each episode, we'll explorethe cutting-edge trends, tools
and strategies driving today'sfinancial landscapes and paving
(00:30):
the way for the future.
With the finance industry at apivotal point, influenced by
groundbreaking innovations, it'smore crucial than ever to
understand how thesetechnological advancements
interact with market dynamics.
In today's episode, we'recovering one of the most talked
(00:56):
about innovations of our timeartificial intelligence.
2023 was the year much of theworld discovered it.
2024 is the year companies arereally starting to use it.
While there's speculation aboutAI's future potential, we're
here to explore its currentimpact and how companies in the
finance industry are harnessingAI and embedding it into their
product suites.
(01:16):
We'll discuss the benefits, thehurdles, any enduring
challenges AI brings to thefinancial landscape.
Joining us to discuss is acompany taking big leaps when it
comes to AI Rance Meshek, ceoof iVest Plus, and Christopher
Mercer, the company's chiefoperating officer.
Ivest is a trading platformdesigned for educators and
(01:36):
retail investors.
Founded to reimagine tradingplatforms, ivest Plus aims to
empower self-directed traderswith innovative tools and
resources.
The platform providescomprehensive trading solutions
that include stocks and options,and advanced data and insights
packaged into user-friendlytechnology.
The company has investedheavily in AI-powered tools,
(01:58):
which Rance and Chris willexplain further.
Gentlemen, welcome to the show.
Speaker 2 (02:02):
Glad to be here.
Thanks for having us.
Speaker 1 (02:04):
Well, why don't we just start at the beginning?
Why don't you give us a littlebit more background on iVest
Plus?
Speaker 3 (02:08):
So iVest was started about 10 years ago now.
It was in 2013.
And the reason I founded thiswas that I had a company I had
sold to TD Ameritrade.
Within a very short period oftime, I became director of
options trading at the firm, andwhat I found was that, not just
at Ameritrade, which is nowpart of Schwab, but all these
(02:36):
brokers, the tools that theygive you are I mean, there's
some good tools, but it's not ascomprehensive as I think it
should be Things like journaling, so you know what's working for
you and what's not.
The depth of information it has.
I mean, their whole goal is getyou to the trade.
Our goal is get you theinformation you need to be able
to trade effectively, and sowhen I left Ameritrade and
founded Ives Plus, we wantedsomething to do exactly that.
(02:58):
So we have best in classjournaling.
We have very comprehensivecompetitive research and
fundamental research oncompanies, great charting and
all that that you'd expect, butthere is a lot more depth than
what a lot of brokers offer togive you a more comprehensive
look at things, and one of thethings we also did with this was
we made it very visual, sopeople don't want to look at
(03:21):
P&Ls and go through all thenumbers, but they want to be
able to get a snapshot of it ina very visual, very quick way to
be able to absorb this.
And that's what we really builton and what we founded Ives for
was to help people that are notin the markets all day be able
to go in, make educateddecisions with easy to
(03:42):
comprehend information and getthem to the trade and then
effectively manage that tradeand then be able to see their
results, and a lot of that isnot available at any of the
brokerage firms and that's whatI wanted to bring to the market
and we've done a great job withit and great adoption through
those additional tools.
Speaker 1 (03:59):
So your company has one of the most comprehensive AI
toolkits I have seen.
Why did you decide so heavilyto invest in AI?
Speaker 2 (04:26):
Well, I think obviously AI is a big flag right
now for everybody, but we sawan opportunity to build that
into certain parts of theplatform.
That would make it a littlemore conversational for people
to use the tools we'd alreadymade, and so we put it on the
project list early this year andbasically got to work on all
these different pieces and kindof integrate them into stuff we
already had.
So it really helped bring a lotto the forefront that we hadn't
really thought through before,make it even easier for people
to use.
Speaker 1 (04:45):
So just to give a little bit more context, can you
please explain what your fourAI tools screener, journal coach
and analyst actually do?
Speaker 3 (04:54):
Absolutely so.
One of the things that we foundthat was a problem with AI was,
if you made it too broad, yousit down and go okay, I'm going
to trade in the markets.
What am I going to do with this?
Right?
So if you just have, like youknow, chat GPT, we're just
talking to it.
We found that people get lostin that.
So what we did was we built fourvery specific tools to really
(05:16):
help people be able to do theiranalysis more effectively and in
a quicker way.
Analysis more effectively andin a quicker way.
So, for example, one of thefirst things that we added was a
competitive analysis on acompany.
So, for example, let's say thatI was going to look at Merck,
(05:36):
and when I look at Merck as apharmaceutical company, I wanted
to know how is it doing?
What's its strengths andweaknesses.
So what we did was we built anAI tool that would and this was
our first entry there's noprompting, there's no discussion
.
It's all us feeding it thequestions, the right information
and then the user being able tosee what that is.
(05:59):
So, for example, on somethinglike Merck, it would tell you
what its strengths are, thingslike you know, it's strong
market cap and so positivechange recently in their pricing
, although, looking at right now, it's saying it's down a little
bit in the last week, right,stable dividends, things like
that.
But it also talks about, youknow, some of the weaknesses
declining prices, you know, highPE ratio, so it's a little
(06:22):
overvalued right now.
Declining income, which isobviously not good.
So then what it would do islook at that, not only give you
your strengths and weaknessesthere, but then it would tell
you other stocks in the industrythat you may want to consider
instead.
So you know, for example, oneof them that came up with was
Johnson Johnson.
That has a better valuation anda stronger investment choice
(06:47):
than Merck.
You know ABBV has a strongdividend and so on.
So you know, then you start tolook at some of this and you
know, you see that, wow, soMerck may not be performing
great, but if I look at some ofthese others, like, for example,
johnson Johnson, and we seethat there's some strong price
improvement on Johnson Johnson,same thing with ABBV.
So what happens is it reallyhelps you, uh, look at a stock,
(07:11):
see what its strengths andweaknesses are and maybe find a
better fit for you based on whatyour outcome is in your
investing.
So that's one of them.
That was the competitiveanalysis, and that was our first
entry into this, where itwasn't.
You know where you're havingthis dialogue part.
It's just.
Here's the answer using AI.
Now, one of the challenges withthis is you want to make sure
(07:34):
the information is accurate andthe training on these things and
how up-to-date they are and allthat.
So what we do is all of theinformation is up-to-date as of
the moment when you go and askthe question.
We actually give it the currentfundamental data and from that
it will generate that report andgive you that information.
(07:54):
So not only have we done a lotof training on it throughout the
whole process, but we also giveit the current information of
all of the companies in itsindustry group for it to be able
to do the analysis on that.
So that's one of the firstthings we did was the
competitive analysis.
Speaker 1 (08:11):
So, rance, let me just stay right there for a
second right.
So you know, I'm seeing anincredible engine.
Where's the fuel?
Where's the data right?
You know, when I'm thinking ofequity trading, I'm thinking of
calendars, I'm thinking ofearnings reports, I'm thinking
of real-time quotes.
You know, how are you poweringthis to give your users the
(08:32):
confidence that the AI has allthat right information as of
this moment?
Speaker 3 (08:37):
So what happens is, you know, in our platform we
have very comprehensivefundamental analysis on
companies, all the way down tolooking at things like the P&L
statements and all that Everyday as we pull that information
in, process it, you know, comeup with a lot of our visuals and
that that we do within theplatform looking at, you know,
(08:59):
key ratios and, you know,comparing it to other companies
within the group and all that.
We take that information everyday and we load that into our
models.
When you ask the question, whenyou hit the button to say, hey,
what's my competitive analysisright now on this, what it's
going to do is it's going tothen send it the updated price
information that it has rightnow.
(09:20):
So it's current, real-timeinformation that it's been given
, plus all the fundamentalthat's loaded every day, right,
so that way we can give itovernight this whole
comprehensive list and thenupdate it with the real-time
price information at the moment,so that what you're looking at
is literally today's data, up toreal-time information about
(09:44):
what's going on with the stocks.
And that's one of thechallenges with this, because if
you look at, like ChatGPT, it'sfrom months ago, right, it's
not current now.
And what we've done is madesure that it had the information
needed right at the moment tobe able to do this.
Speaker 1 (09:59):
Wow, all right.
Well, take me through the othertools, all right.
Speaker 3 (10:03):
So one of one of the other things that we did then
was we thought you know what,let's go with a financial coach,
a chat bot kind of thing, butkeep it within the bounds of
financial.
So you know, if you were to askthe question, should I vote red
or blue in this election?
It's not going to give you thatinformation, but what it's
(10:25):
going to do is tell you maybewhat industry groups work better
under Republican, what industrygroups work better under
Democrat rule, right?
So it gives you informationspecifically about trading.
But in this we did open this toa broader chat capability and
so, for example, let's just sayyou know, if I ask it something
like how do I structure a bullput spread and when should I
(10:47):
apply this strategy, it willtell me then exactly how I'm
going to structure the trade,how to put it together, and then
will tell me when I wouldconsider applying that in the
market, and then from that Icould ask additional questions,
have back and forth dialoguewith this on how I'd want to be
able to do that.
And what we also found shortlyafter launching the chatbot was
(11:08):
people wanted to go back toprevious conversations and be
able to go into those andfurther that particular
conversation.
So if I were to take one youknow I asked the other day about
, you know, give me 20 goodstocks under 50 bucks, and you
know it came up with a list ofthose.
And then what we also do withthis is we show you a list of
what stocks it came up with soyou can just click on it and go
(11:30):
to further analysis on them ifyou want to.
But the fact that you can thenkeep those conversations going,
we allow you to keep up to thelast 28 days of use.
So if I had something I didthree months ago but I've been
going back to that conversation,it's always there for me to be
able to continue on.
(11:50):
So I can take, let's say, backto that bull put spread right
that particular option strategythat I don't quite get.
I can ask, let's say, back tothat bull put spread right that
particular option strategy thatI don't quite get.
I can ask more questions aboutit and dive deeper into it.
So that was the second one wedid, which was a broader chatbot
.
Speaker 1 (12:05):
I'm going to dive in with a follow-up.
One of the things I noticed asyou're kind of going through it,
your prompts were very simpleWas that by design I mean as a
chat, gptt user and co-pilot andwhatnot.
You know, I've been honing myprompt engineering skills, but
this seems a little bit morestraightforward.
Speaker 3 (12:25):
Right, you know, like a real simple one here is how
do I structure a bullpup spread?
Right, you know it's likethere's no prompting to that.
So here's what we did with this, and I think this is one of the
secret sauces to this.
You know, it's great thatyou've been using these and
you've learned that how you askyour question is incredibly
important, right, and you knowhow you build that prompt.
(12:49):
So what we did for the user iswe took that need for that
knowledge off of them and webuilt it into our backend.
So when you ask a simplequestion like you know, how do I
structure a bullfoot spreadBehind the scenes, the prompt
that's actually being asked of.
It is a much more comprehensiveprompt.
It's your topic, framed in away that we know will get you
(13:11):
the right type of answer backRight, framed in a way that we
know will get you the right typeof answer back Right.
So it again makes that.
Yeah, I've been using ChatGPTfor quite a few things.
I'm building a closet systemright now and how to do certain
things I'm asking it questionson and how you.
The prompting is so importanton it and we found that if we
(13:33):
didn't hone in on that itreduced the value to the user,
because then they'd have tolearn this skill of prompt right
, and what we did was we tookthat away.
Let them ask in a simple way.
We then take that in our modelsand how to structure it.
So the prompt is asked in avery good way.
Plus, we've also done a lot oftraining of the AI models that
(13:53):
we're using to make sure thatthe combination of the prompt
and the training gives themsolid answers.
Speaker 1 (13:59):
Got it, wow, okay, what's next?
Speaker 3 (14:02):
All right.
So here's one of the thingsthat we found with this.
When we went to this, thatissue of too open-ended came up.
Right.
It was.
We found that people were.
They liked it, but how do Ireally use this in my trading?
And so then what we did was weworked on that and found that,
you know, we built it into ourscreener.
(14:23):
Now here's the thing about ourscreener system.
If you look at the screenersystem that we have, we
literally have somewhere around350 different data points to
choose from.
It's a tremendous amount ofdata.
So somebody that is relativelynew to the markets doesn't pour
(14:45):
in the numbers all day long andall that stuff.
They don't even know what tochoose.
Right, yeah, I want stocks overbetween $50 and $100.
Okay, but when you start to getinto financial strength and
what that means and all that,that became a lot more
challenging for our users to do.
So what we did was we built anAI bot specifically around our
(15:05):
screener where you can ask itquestions and what it will do is
we'll decide what the fieldsare appropriate for, what your
question is and what range itshould be.
So, for example, you know,let's just go with.
You know I'm looking to docovered calls on stocks with
solid fundamentals that pay agood dividend.
(15:26):
And so then it'll tell you,it'll come back with an answer
as to what it's going to lookfor and why, and then it will
come up with a list of theactual fields and the ranges
that it should use in thosefields.
And then what it'll do isdisplay how many stocks fall
into those different categories.
And so when I did this, thefirst question when I asked that
(15:48):
question about covered callswith good fundamentals that
could pay a good premium, I hadlike 450 stocks.
That was like too many.
So now I'm back to this backand forth dialogue specific to
my screener and it's like, hey,this is still too many stocks.
Let's add some fundamentalfilters to make sure it's a good
, strong company, but also wedon't want something that's too
(16:09):
volatile.
And then it would come backwith what you know, what it
thinks it's needed there, and ithones in even more.
And then you know I'd say it'sstill a little bit more.
So then I said, hey, one of thethings that happened when I did
this was there was an $800stock on the list and I wasn't
looking to buy 100 shares of an$800 stock and if you're going
to do a covered call, you kindof need to do it in 100s or
(16:29):
increments.
So I said, hey, let's make sureit's under $150.
And so it added that to thelist.
And what started out with?
You know, if we look at ourwhole universe of stocks, we
have about 25,000 stocks.
We have about 12,000 fullylisted stocks.
The rest of them are likebulletin board over the counter
stuff.
So with over 12,000 stocksthrough this, it shows what the
(16:52):
fields were, what the paramsshould be for those fields.
And now I'm down to like 15stocks right out of this by
those simple questions.
And again, notice that to yourthing about prompting.
I'm not having to know how toconstruct a prompt.
We've taken care of that forthe user.
We make it a really simplething.
Hey, I'm looking to do coveredcalls and I want good premium.
(17:12):
Boom.
You see, you still have too manystocks.
What do you want to hone it inon?
You know you still have toomany stocks.
What do you want to hone it inon?
You know, hey, let's make surethey're good.
You know solid fundamentals,but I don't want something too
volatile.
Right, that's what I did in myexample here and it hones it in.
And then I've got my 15 stockslisted here and, by the way,
once I've done that, once I canalso save that particular
(17:34):
screener for use anytime I wantto right.
But so then I've got my stockson here that I'm working on and
just as a quick little examplehere, if I went with I don't
know, let's just go with one ofthese here we have, aem was one
of the first ones on our listand if I looked at this and went
to do a covered call on thatparticular stock, it would go in
(17:56):
.
Our system will do the analysison it, come up with this what
the structure should be.
And here I have a stock that'saround $72.
If I were to do a $75, and ifyour users know that covered
callers are selling the rightfor somebody to buy away from
you at 75 when the stock's at 72, you know I'm set up for about
a 7% return in about six weeks,right.
(18:19):
And you know, if I want alittle bit more room for it to
move up, I can, you know, adjustthat accordingly.
But you know now I'm up toabout a 12% return in about six
weeks if it goes up to $80,right.
So what it does is, you know,honed in on the list.
Then I go start to do myresearch on that list of stocks
and see how it plays out.
But the fact that you can nowdo this with very simple, just
(18:44):
little conversation, withouthaving to build big prompts, all
that stuff or have any cluewhat all these different things
even mean.
In this example, we have over adozen criteria that it came up
with and you didn't have to knowwhat that stuff even meant.
You just look at hey, I got 15stocks.
Okay, great, that's a goodnumber.
Let me go look at this and seehow they play out.
(19:05):
And, and you know, justsubstantially short circuits or
shortens the length of time ittakes you to get to your answers
, and reduces the amount ofinformation and understanding
you need about all thisdifferent data to be able to do
it.
The AI decides that for you, soit just simplifies the process
on this substantially to be ableto really kind of hone in on
(19:30):
finding what you want to find.
And then there's one more thatI got to share with you, because
this is such a powerful pieceof this In our platform in
general, one of the thingsthat's really, really powerful
about it is our journalingcapability, and if anybody is
doing you know for anybody yourlisteners that might do options
(19:51):
trading.
Options trading in journaling isa real challenging thing and
we've done a lot to build outour journaling capability around
stock and around futures andalso around options.
But one of the things thathappened so you get to see
what's working, what's not, whatyour returns are on different
things and all that but what wefound was that you know you
(20:16):
still have to go pour throughthis and say, ok, why did this
work?
Why did this not work, whatever?
Well, what we did was we nowbuilt in this was our latest one
where we'll take your tradinghistory of, and you can do this,
you can hone it in, you canlook at a particular sector, you
could look at certain dateranges.
(20:37):
However, you want to hone inthe journal, right?
So you use the filters we havefor the journal, you get at what
you want to look at, then yousend it over to AI and then what
will happen is it will analyzeyour trading not general, your
specific behavior and tell youwhat your strengths are, what
your areas for improvement are,and so, for example, one I've
(20:58):
just run here is I'm doingreally good with bull call
spread strategies.
I'm doing really good onprofits when I do my stock
trades not just opposition, butstock itself.
One of the things I happen to do, though, in my areas for
improvement is sometimes I letthings that aren't working out
(21:19):
real well run too long.
I don't cut my losses, I don'tclose a trade out and I'm hoping
it.
You know that old saying thatnothing turns a short-term trade
into a long-term investmentquicker than the stock going in
the wrong direction, right, andI find myself sometimes having
that issue.
And so it's telling me hey, youknow, I'm letting some things
(21:39):
run through options, you knowthrough options expiration, and
it's biting me and I'm havingsome trades that you know that I
didn't close them out in time,right, and so it gives me, it
tells me areas of improvement,tells what I'm doing well, what
my strengths are, areas forimprovement, some general
(21:59):
observation.
So, just as an example, here inmy portfolio, it's telling me
hey, I've got a good mix ofsuccessful trades, profitable
outcomes through this.
I'm using diversificationacross different strategies and
so on, so I'm doing well in thatarea.
But it's telling me that, hey,you need to focus on three key
(22:20):
areas to improve your trading.
You need to focus on completedtrades, closing out trades that
aren't working for you.
Don't let them run to the end.
Maybe you can do a little bitbetter job at active trade
management.
And if something's not working,you know, review it, maybe
adjust it or whatever, but don'tjust let it hang out there.
Right?
(22:40):
So it's telling me areas that Ican focus on to improve my
trading and but but even if youdon't, if you just say, hey,
what's working well and you domore of what's working well and
stop doing what's not workingwell, you're going to do better
in your trading.
But it goes beyond that.
It not only tells you whatyou're doing well and what
you're not doing well.
It also will tell you whatareas you should focus on to
(23:04):
improve on your trading results.
Speaker 1 (23:12):
Thank you for listening to part one of this
episode on AI and investing withRance Meschick, the CEO of
iVest Plus, and COO Chris Mercer.
Stay tuned next week as werelease part two of this
exciting conversation.
Welcome to Trading Tomorrow Navigating Trends in
Capital Markets the podcastwhere we deep dive into
technologies reshaping the worldof capital markets.
I'm your host, jim Jockle, aveteran of the finance industry
with a passion for thecomplexities of financial
technologies and market trends.
In each episode, we'll explorethe cutting-edge trends, tools
and strategies driving today'sfinancial landscapes and paving
(00:30):
the way for the future.
With the finance industry at apivotal point, influenced by
groundbreaking innovations, it'smore crucial than ever to
understand how thesetechnological advancements
interact with market dynamics.
In today's episode, we'recovering one of the most talked
(00:56):
about innovations of our timeartificial intelligence.
2023 was the year much of theworld discovered it.
2024 is the year companies arereally starting to use it.
While there's speculation aboutAI's future potential, we're
here to explore its currentimpact and how companies in the
finance industry are harnessingAI and embedding it into their
product suites.
(01:16):
We'll discuss the benefits, thehurdles, any enduring
challenges AI brings to thefinancial landscape.
Joining us to discuss is acompany taking big leaps when it
comes to AI Rance Meshek, ceoof iVest Plus, and Christopher
Mercer, the company's chiefoperating officer.
Ivest is a trading platformdesigned for educators and
(01:36):
retail investors.
Founded to reimagine tradingplatforms, ivest Plus aims to
empower self-directed traderswith innovative tools and
resources.
The platform providescomprehensive trading solutions
that include stocks and options,and advanced data and insights
packaged into user-friendlytechnology.
The company has investedheavily in AI-powered tools,
(01:58):
which Rance and Chris willexplain further.
Gentlemen, welcome to the show.
Speaker 2 (02:02):
Glad to be here.
Thanks for having us.
Speaker 1 (02:04):
Well, why don't we just start at the beginning?
Why don't you give us a littlebit more background on iVest
Plus?
Speaker 3 (02:08):
So iVest was started about 10 years ago now.
It was in 2013.
And the reason I founded thiswas that I had a company I had
sold to TD Ameritrade.
Within a very short period oftime, I became director of
options trading at the firm, andwhat I found was that, not just
at Ameritrade, which is nowpart of Schwab, but all these
(02:36):
brokers, the tools that theygive you are I mean, there's
some good tools, but it's not ascomprehensive as I think it
should be Things like journaling, so you know what's working for
you and what's not.
The depth of information it has.
I mean, their whole goal is getyou to the trade.
Our goal is get you theinformation you need to be able
to trade effectively, and sowhen I left Ameritrade and
founded Ives Plus, we wantedsomething to do exactly that.
(02:58):
So we have best in classjournaling.
We have very comprehensivecompetitive research and
fundamental research oncompanies, great charting and
all that that you'd expect, butthere is a lot more depth than
what a lot of brokers offer togive you a more comprehensive
look at things, and one of thethings we also did with this was
we made it very visual, sopeople don't want to look at
(03:21):
P&Ls and go through all thenumbers, but they want to be
able to get a snapshot of it ina very visual, very quick way to
be able to absorb this.
And that's what we really builton and what we founded Ives for
was to help people that are notin the markets all day be able
to go in, make educateddecisions with easy to
(03:42):
comprehend information and getthem to the trade and then
effectively manage that tradeand then be able to see their
results, and a lot of that isnot available at any of the
brokerage firms and that's whatI wanted to bring to the market
and we've done a great job withit and great adoption through
those additional tools.
Speaker 1 (03:59):
So your company has one of the most comprehensive AI
toolkits I have seen.
Why did you decide so heavilyto invest in AI?
Speaker 2 (04:26):
Well, I think obviously AI is a big flag right
now for everybody, but we sawan opportunity to build that
into certain parts of theplatform.
That would make it a littlemore conversational for people
to use the tools we'd alreadymade, and so we put it on the
project list early this year andbasically got to work on all
these different pieces and kindof integrate them into stuff we
already had.
So it really helped bring a lotto the forefront that we hadn't
really thought through before,make it even easier for people
to use.
Speaker 1 (04:45):
So just to give a little bit more context, can you
please explain what your fourAI tools screener, journal coach
and analyst actually do?
Speaker 3 (04:54):
Absolutely so.
One of the things that we foundthat was a problem with AI was,
if you made it too broad, yousit down and go okay, I'm going
to trade in the markets.
What am I going to do with this?
Right?
So if you just have, like youknow, chat GPT, we're just
talking to it.
We found that people get lostin that.
So what we did was we built fourvery specific tools to really
(05:16):
help people be able to do theiranalysis more effectively and in
a quicker way.
Analysis more effectively andin a quicker way.
So, for example, one of thefirst things that we added was a
competitive analysis on acompany.
So, for example, let's say thatI was going to look at Merck,
(05:36):
and when I look at Merck as apharmaceutical company, I wanted
to know how is it doing?
What's its strengths andweaknesses.
So what we did was we built anAI tool that would and this was
our first entry there's noprompting, there's no discussion
.
It's all us feeding it thequestions, the right information
and then the user being able tosee what that is.
(05:59):
So, for example, on somethinglike Merck, it would tell you
what its strengths are, thingslike you know, it's strong
market cap and so positivechange recently in their pricing
, although, looking at right now, it's saying it's down a little
bit in the last week, right,stable dividends, things like
that.
But it also talks about, youknow, some of the weaknesses
declining prices, you know, highPE ratio, so it's a little
(06:22):
overvalued right now.
Declining income, which isobviously not good.
So then what it would do islook at that, not only give you
your strengths and weaknessesthere, but then it would tell
you other stocks in the industrythat you may want to consider
instead.
So you know, for example, oneof them that came up with was
Johnson Johnson.
That has a better valuation anda stronger investment choice
(06:47):
than Merck.
You know ABBV has a strongdividend and so on.
So you know, then you start tolook at some of this and you
know, you see that, wow, soMerck may not be performing
great, but if I look at some ofthese others, like, for example,
johnson Johnson, and we seethat there's some strong price
improvement on Johnson Johnson,same thing with ABBV.
So what happens is it reallyhelps you, uh, look at a stock,
(07:11):
see what its strengths andweaknesses are and maybe find a
better fit for you based on whatyour outcome is in your
investing.
So that's one of them.
That was the competitiveanalysis, and that was our first
entry into this, where itwasn't.
You know where you're havingthis dialogue part.
It's just.
Here's the answer using AI.
Now, one of the challenges withthis is you want to make sure
(07:34):
the information is accurate andthe training on these things and
how up-to-date they are and allthat.
So what we do is all of theinformation is up-to-date as of
the moment when you go and askthe question.
We actually give it the currentfundamental data and from that
it will generate that report andgive you that information.
(07:54):
So not only have we done a lotof training on it throughout the
whole process, but we also giveit the current information of
all of the companies in itsindustry group for it to be able
to do the analysis on that.
So that's one of the firstthings we did was the
competitive analysis.
Speaker 1 (08:11):
So, rance, let me just stay right there for a
second right.
So you know, I'm seeing anincredible engine.
Where's the fuel?
Where's the data right?
You know, when I'm thinking ofequity trading, I'm thinking of
calendars, I'm thinking ofearnings reports, I'm thinking
of real-time quotes.
You know, how are you poweringthis to give your users the
(08:32):
confidence that the AI has allthat right information as of
this moment?
Speaker 3 (08:37):
So what happens is, you know, in our platform we
have very comprehensivefundamental analysis on
companies, all the way down tolooking at things like the P&L
statements and all that Everyday as we pull that information
in, process it, you know, comeup with a lot of our visuals and
that that we do within theplatform looking at, you know,
(08:59):
key ratios and, you know,comparing it to other companies
within the group and all that.
We take that information everyday and we load that into our
models.
When you ask the question, whenyou hit the button to say, hey,
what's my competitive analysisright now on this, what it's
going to do is it's going tothen send it the updated price
information that it has rightnow.
(09:20):
So it's current, real-timeinformation that it's been given
, plus all the fundamentalthat's loaded every day, right,
so that way we can give itovernight this whole
comprehensive list and thenupdate it with the real-time
price information at the moment,so that what you're looking at
is literally today's data, up toreal-time information about
(09:44):
what's going on with the stocks.
And that's one of thechallenges with this, because if
you look at, like ChatGPT, it'sfrom months ago, right, it's
not current now.
And what we've done is madesure that it had the information
needed right at the moment tobe able to do this.
Speaker 1 (09:59):
Wow, all right.
Well, take me through the othertools, all right.
Speaker 3 (10:03):
So one of one of the other things that we did then
was we thought you know what,let's go with a financial coach,
a chat bot kind of thing, butkeep it within the bounds of
financial.
So you know, if you were to askthe question, should I vote red
or blue in this election?
It's not going to give you thatinformation, but what it's
(10:25):
going to do is tell you maybewhat industry groups work better
under Republican, what industrygroups work better under
Democrat rule, right?
So it gives you informationspecifically about trading.
But in this we did open this toa broader chat capability and
so, for example, let's just sayyou know, if I ask it something
like how do I structure a bullput spread and when should I
(10:47):
apply this strategy, it willtell me then exactly how I'm
going to structure the trade,how to put it together, and then
will tell me when I wouldconsider applying that in the
market, and then from that Icould ask additional questions,
have back and forth dialoguewith this on how I'd want to be
able to do that.
And what we also found shortlyafter launching the chatbot was
(11:08):
people wanted to go back toprevious conversations and be
able to go into those andfurther that particular
conversation.
So if I were to take one youknow I asked the other day about
, you know, give me 20 goodstocks under 50 bucks, and you
know it came up with a list ofthose.
And then what we also do withthis is we show you a list of
what stocks it came up with soyou can just click on it and go
(11:30):
to further analysis on them ifyou want to.
But the fact that you can thenkeep those conversations going,
we allow you to keep up to thelast 28 days of use.
So if I had something I didthree months ago but I've been
going back to that conversation,it's always there for me to be
able to continue on.
(11:50):
So I can take, let's say, backto that bull put spread right
that particular option strategythat I don't quite get.
I can ask, let's say, back tothat bull put spread right that
particular option strategy thatI don't quite get.
I can ask more questions aboutit and dive deeper into it.
So that was the second one wedid, which was a broader chatbot
.
Speaker 1 (12:05):
I'm going to dive in with a follow-up.
One of the things I noticed asyou're kind of going through it,
your prompts were very simpleWas that by design I mean as a
chat, gptt user and co-pilot andwhatnot.
You know, I've been honing myprompt engineering skills, but
this seems a little bit morestraightforward.
Speaker 3 (12:25):
Right, you know, like a real simple one here is how
do I structure a bullpup spread?
Right, you know it's likethere's no prompting to that.
So here's what we did with this, and I think this is one of the
secret sauces to this.
You know, it's great thatyou've been using these and
you've learned that how you askyour question is incredibly
important, right, and you knowhow you build that prompt.
(12:49):
So what we did for the user iswe took that need for that
knowledge off of them and webuilt it into our backend.
So when you ask a simplequestion like you know, how do I
structure a bullfoot spreadBehind the scenes, the prompt
that's actually being asked of.
It is a much more comprehensiveprompt.
It's your topic, framed in away that we know will get you
(13:11):
the right type of answer backRight, framed in a way that we
know will get you the right typeof answer back Right.
So it again makes that.
Yeah, I've been using ChatGPTfor quite a few things.
I'm building a closet systemright now and how to do certain
things I'm asking it questionson and how you.
The prompting is so importanton it and we found that if we
(13:33):
didn't hone in on that itreduced the value to the user,
because then they'd have tolearn this skill of prompt right
, and what we did was we tookthat away.
Let them ask in a simple way.
We then take that in our modelsand how to structure it.
So the prompt is asked in avery good way.
Plus, we've also done a lot oftraining of the AI models that
(13:53):
we're using to make sure thatthe combination of the prompt
and the training gives themsolid answers.
Speaker 1 (13:59):
Got it, wow, okay, what's next?
Speaker 3 (14:02):
All right.
So here's one of the thingsthat we found with this.
When we went to this, thatissue of too open-ended came up.
Right.
It was.
We found that people were.
They liked it, but how do Ireally use this in my trading?
And so then what we did was weworked on that and found that,
you know, we built it into ourscreener.
(14:23):
Now here's the thing about ourscreener system.
If you look at the screenersystem that we have, we
literally have somewhere around350 different data points to
choose from.
It's a tremendous amount ofdata.
So somebody that is relativelynew to the markets doesn't pour
(14:45):
in the numbers all day long andall that stuff.
They don't even know what tochoose.
Right, yeah, I want stocks overbetween $50 and $100.
Okay, but when you start to getinto financial strength and
what that means and all that,that became a lot more
challenging for our users to do.
So what we did was we built anAI bot specifically around our
(15:05):
screener where you can ask itquestions and what it will do is
we'll decide what the fieldsare appropriate for, what your
question is and what range itshould be.
So, for example, you know,let's just go with.
You know I'm looking to docovered calls on stocks with
solid fundamentals that pay agood dividend.
(15:26):
And so then it'll tell you,it'll come back with an answer
as to what it's going to lookfor and why, and then it will
come up with a list of theactual fields and the ranges
that it should use in thosefields.
And then what it'll do isdisplay how many stocks fall
into those different categories.
And so when I did this, thefirst question when I asked that
(15:48):
question about covered callswith good fundamentals that
could pay a good premium, I hadlike 450 stocks.
That was like too many.
So now I'm back to this backand forth dialogue specific to
my screener and it's like, hey,this is still too many stocks.
Let's add some fundamentalfilters to make sure it's a good
, strong company, but also wedon't want something that's too
(16:09):
volatile.
And then it would come backwith what you know, what it
thinks it's needed there, and ithones in even more.
And then you know I'd say it'sstill a little bit more.
So then I said, hey, one of thethings that happened when I did
this was there was an $800stock on the list and I wasn't
looking to buy 100 shares of an$800 stock and if you're going
to do a covered call, you kindof need to do it in 100s or
(16:29):
increments.
So I said, hey, let's make sureit's under $150.
And so it added that to thelist.
And what started out with?
You know, if we look at ourwhole universe of stocks, we
have about 25,000 stocks.
We have about 12,000 fullylisted stocks.
The rest of them are likebulletin board over the counter
stuff.
So with over 12,000 stocksthrough this, it shows what the
(16:52):
fields were, what the paramsshould be for those fields.
And now I'm down to like 15stocks right out of this by
those simple questions.
And again, notice that to yourthing about prompting.
I'm not having to know how toconstruct a prompt.
We've taken care of that forthe user.
We make it a really simplething.
Hey, I'm looking to do coveredcalls and I want good premium.
(17:12):
Boom.
You see, you still have too manystocks.
What do you want to hone it inon?
You know you still have toomany stocks.
What do you want to hone it inon?
You know, hey, let's make surethey're good.
You know solid fundamentals,but I don't want something too
volatile.
Right, that's what I did in myexample here and it hones it in.
And then I've got my 15 stockslisted here and, by the way,
once I've done that, once I canalso save that particular
(17:34):
screener for use anytime I wantto right.
But so then I've got my stockson here that I'm working on and
just as a quick little examplehere, if I went with I don't
know, let's just go with one ofthese here we have, aem was one
of the first ones on our listand if I looked at this and went
to do a covered call on thatparticular stock, it would go in
(17:56):
.
Our system will do the analysison it, come up with this what
the structure should be.
And here I have a stock that'saround $72.
If I were to do a $75, and ifyour users know that covered
callers are selling the rightfor somebody to buy away from
you at 75 when the stock's at 72, you know I'm set up for about
a 7% return in about six weeks,right.
(18:19):
And you know, if I want alittle bit more room for it to
move up, I can, you know, adjustthat accordingly.
But you know now I'm up toabout a 12% return in about six
weeks if it goes up to $80,right.
So what it does is, you know,honed in on the list.
Then I go start to do myresearch on that list of stocks
and see how it plays out.
But the fact that you can nowdo this with very simple, just
(18:44):
little conversation, withouthaving to build big prompts, all
that stuff or have any cluewhat all these different things
even mean.
In this example, we have over adozen criteria that it came up
with and you didn't have to knowwhat that stuff even meant.
You just look at hey, I got 15stocks.
Okay, great, that's a goodnumber.
Let me go look at this and seehow they play out.
(19:05):
And, and you know, justsubstantially short circuits or
shortens the length of time ittakes you to get to your answers
, and reduces the amount ofinformation and understanding
you need about all thisdifferent data to be able to do
it.
The AI decides that for you, soit just simplifies the process
on this substantially to be ableto really kind of hone in on
(19:30):
finding what you want to find.
And then there's one more thatI got to share with you, because
this is such a powerful pieceof this In our platform in
general, one of the thingsthat's really, really powerful
about it is our journalingcapability, and if anybody is
doing you know for anybody yourlisteners that might do options
(19:51):
trading.
Options trading in journaling isa real challenging thing and
we've done a lot to build outour journaling capability around
stock and around futures andalso around options.
But one of the things thathappened so you get to see
what's working, what's not, whatyour returns are on different
things and all that but what wefound was that you know you
(20:16):
still have to go pour throughthis and say, ok, why did this
work?
Why did this not work, whatever?
Well, what we did was we nowbuilt in this was our latest one
where we'll take your tradinghistory of, and you can do this,
you can hone it in, you canlook at a particular sector, you
could look at certain dateranges.
(20:37):
However, you want to hone inthe journal, right?
So you use the filters we havefor the journal, you get at what
you want to look at, then yousend it over to AI and then what
will happen is it will analyzeyour trading not general, your
specific behavior and tell youwhat your strengths are, what
your areas for improvement are,and so, for example, one I've
(20:58):
just run here is I'm doingreally good with bull call
spread strategies.
I'm doing really good onprofits when I do my stock
trades not just opposition, butstock itself.
One of the things I happen to do, though, in my areas for
improvement is sometimes I letthings that aren't working out
(21:19):
real well run too long.
I don't cut my losses, I don'tclose a trade out and I'm hoping
it.
You know that old saying thatnothing turns a short-term trade
into a long-term investmentquicker than the stock going in
the wrong direction, right, andI find myself sometimes having
that issue.
And so it's telling me hey, youknow, I'm letting some things
(21:39):
run through options, you knowthrough options expiration, and
it's biting me and I'm havingsome trades that you know that I
didn't close them out in time,right, and so it gives me, it
tells me areas of improvement,tells what I'm doing well, what
my strengths are, areas forimprovement, some general
(21:59):
observation.
So, just as an example, here inmy portfolio, it's telling me
hey, I've got a good mix ofsuccessful trades, profitable
outcomes through this.
I'm using diversificationacross different strategies and
so on, so I'm doing well in thatarea.
But it's telling me that, hey,you need to focus on three key
(22:20):
areas to improve your trading.
You need to focus on completedtrades, closing out trades that
aren't working for you.
Don't let them run to the end.
Maybe you can do a little bitbetter job at active trade
management.
And if something's not working,you know, review it, maybe
adjust it or whatever, but don'tjust let it hang out there.
Right?
(22:40):
So it's telling me areas that Ican focus on to improve my
trading and but but even if youdon't, if you just say, hey,
what's working well and you domore of what's working well and
stop doing what's not workingwell, you're going to do better
in your trading.
But it goes beyond that.
It not only tells you whatyou're doing well and what
you're not doing well.
It also will tell you whatareas you should focus on to
(23:04):
improve on your trading results.
Speaker 1 (23:12):
Thank you for listening to part one of this
episode on AI and investing withRance Meschick, the CEO of
iVest Plus, and COO Chris Mercer.
Stay tuned next week as werelease part two of this
exciting conversation.
Popular Podcasts
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com