Published on

BA04-1. [Novel] Probability in Saigon — The Day Data Beat Intuition (Part 1)

BA04-1. [Novel] Probability in Saigon — The Day Data Beat Intuition (Part 1)

Probability in Saigon — The Day Data Beat Intuition

"Probability does not lie. It simply states 'I don't know yet' honestly when evidence is lacking."


Prologue: A City of 38 Degrees, A Game of 10 Billion

February 2026, Ho Chi Minh City.

Dozens of cranes pierced the sky above the Saigon River. With a temperature of 38 degrees Celsius and 85% humidity, just walking down the street made shirts cling to backs. In the heart of District 2 (Thu Duc City) of this sweltering city, a pre-sales war for the 58-story ultra-high-rise condominium, 'Saigon River Sky,' was unfolding.

A total of 480 units. The average pre-sale price per unit was 4.5 billion VND (approx. 250 million KRW). The total project value was about 120 billion KRW.

The top three real estate agencies in Ho Chi Minh City were clashing over the exclusive sales agency rights for this project. The agency commission alone amounted to 3% of total revenue — roughly 3.6 billion KRW. To seize this 3.6 billion KRW, each firm deployed its best sales team.

The ace of Nam Viet Realty, Nguyễn Văn Tuấn. A 38-year-old veteran known in the industry as the 'Wolf of Saigon'. 15 years of experience, with cumulative pre-sales of 3,200 units. His weapon was intuition. The look in a customer's eyes, the firmness of a handshake, where their gaze lingered during a conversation — he read it all. Tuấn's closing rate was 2.3 times the industry average. If you didn't know him in the Ho Chi Minh real estate industry, you were living under a rock.

And standing opposite Tuấn was the newly appointed team leader of 'SkyLink Property', an unfamiliar startup agency: Park Jun-hyuk. A 34-year-old man who ran an IT startup in Korea before moving to Ho Chi Minh two years ago. His real estate experience was a mere year and a half. His weapon was the blue dashboard on his laptop screen — EXAWin.

"Brother Tuấn, we're going to be number one again this time, right?"

A rookie at Nam Viet Realty asked. Sitting on his motorbike, Tuấn took a sip of peppermint coffee and looked at SkyLink's office across the street. Through the glass window, Park Jun-hyuk's team members were sitting in front of their laptops, typing away furiously.

"Those guys... they think they can sell real estate with computers."

Tuấn snorted. He was confident that his 15 years of intuition was better than any software.

That was a mistake.

Chapter 1: The First Battle — The VIP List

1.1 Tuấn's Method: Golf Courses and Bar Lounges

The first stage of the pre-sales war is securing a VIP prospect list. Saigon River Sky is a premium condo priced at 4555 million VND per pyeong (approx. 2.53.1 million KRW). The target audience was clear:

  1. Korean investors residing in Ho Chi Minh City (Abundant cash, purchasing for investment)
  2. Vietnam's newly wealthy class (CEOs of IT, finance, and manufacturing)
  3. Foreign investors from Singapore, Taiwan, and Hong Kong (Portfolio diversification)
  4. Local expat business owners (For both residence and investment)

Tuấn moved in his own way. He played rounds with successful business owners at 'Tan City Golf Club', a premium golf course in District 7, and drank whiskey with Korean investors at a rooftop bar on Đồng Khởi street in District 1. He handed out business cards and spoke in hushed tones.

"Brother, you know about the new one going up on the riverbank in District 2, Thu Duc, right? 58 stories. Direct river view. If you don't grab it now, it'll be gone."

Tuấn's VIP list reached 87 people in just two weeks. It was mostly a 'network-based' list built at golf courses and drinking sessions. He differentiated who among these 87 people were real buyers based on his gut feeling (感).

"This person's eyes are different. He's serious."

"This person is just gathering information. A waste of time."

Tuấn categorized them into 23 A-grade (will definitely buy), 41 B-grade (has potential), and 23 C-grade (just interested). The criteria? Intuition accumulated over 15 years.

1.2 Park Jun-hyuk's Method: Data and Signals

At the same time, Park Jun-hyuk of SkyLink Property was staring at the EXAWin dashboard in his office.

"Team Leader, shouldn't we be going to the golf course too?"

Team member Lê asked. Park Jun-hyuk shook his head.

"How do you know if the person you meet at the golf course is a real buyer? By feeling?"

Park Jun-hyuk created the Saigon River Sky Pre-sales Project in EXAWin. He set the values like this:

ParameterValueReason
Prior Alpha (α₀)1.0Startup agency, no prior track record — A humble start
Prior Beta (β₀)1.5Base failure weight set slightly higher due to fierce competition
Silence Gap5 DaysReal estate customers cool off quickly — 5 days of silence is a risk
Silence Interval3 DaysPenalty accumulates every 3 days
Threshold (T)0.75If 75% or higher, the contract can be pushed forward

"Instead of golf courses, we collect leads through online ads, real estate portals, expat communities, and business networks. What matters is not the number of people you meet, but what you record after meeting them."

Park Jun-hyuk's team recorded every interaction with potential customers contacted over the two weeks into EXAWin. Online inquiries, phone consultations, model house visits, 1:1 meetings — they entered signals at every touchpoint.

  • "Requested a site visit" → Weak Affirmation (WA 2.0)
  • "Specific inquiry about ROI" → Strong Affirmation (SA 5.0)
  • "Looking at other projects as well" → Weak Negation (WN 2.0)
  • "Price is too high" → Strong Negation (SN 5.0)
  • "Will discuss with family" → Weak Negation (WN 1.0)
  • "Inquired about how to prepare the down payment" → Strong Affirmation (SA 5.0)

Two weeks later, Park Jun-hyuk's list had 62 people. 25 fewer than Tuấn. But the difference lay in the quality.

On the EXAWin dashboard, the P(Win) — the probability of closing the contract — for each of the 62 people was displayed in real-time. And they were categorized as 🌱Early, 🌿Growing, or 🌳Mature based on evidence maturity.

"Lê, look at this list. There are 8 people with a P(Win) over 70%. Focus on these 8."

"What about the other 54?"

"The system will track them automatically. If silence penalties accumulate, an automatic follow-up notification will pop up. We focus our energy where the probability is high."


Chapter 2: The First Contract Competition — President Phan

2.1 Same Customer, Different Approaches

Phan Văn Minh, 45 years old. A newly wealthy individual running a logistics company in Ho Chi Minh City. Assets of approximately 30 billion VND (1.7 billion KRW). A wife and two college-aged daughters. Currently residing in Phú Mỹ Hưng, District 7, and looking for 1~2 condos for investment.

President Phan contacted both sides. He was contacted by Tuấn, and he also submitted an online inquiry to Park Jun-hyuk.

Tuấn's Record (Notepad):

"Phan Minh. Logistics. Investment purpose. Interest level A-grade. Potential for 2 units. Dinner scheduled."

Park Jun-hyuk's Record (EXAWin Activity Log):
  • Activity: Phone consultation 30 mins
  • Stage: Discovery (SWV = 1.0)
  • Signal: WA 2.0 (Intention to visit model house), WA 2.0 (Mentioned investment purpose)
  • Summary: Logistics company president, interested in 1~2 units for investment. Budget unclear. Lives in District 7.

The P(Win) after the first meeting with President Phan calculated by EXAWin:

α1=1.0+1.0×2.0=3.0\alpha_1 = 1.0 + 1.0 \times 2.0 = 3.0 β1=1.5+0=1.5\beta_1 = 1.5 + 0 = 1.5 P1(Win)=3.03.0+1.5=66.7%P_1(Win) = \frac{3.0}{3.0 + 1.5} = 66.7\%

"66.7%. Not bad, but still in the Early stage. We need more evidence."

2.2 Model House Meeting — Tuấn vs Park Jun-hyuk

Three days later, President Phan visited the model house. Coincidentally, Tuấn and Park Jun-hyuk ran into President Phan at the same model house on the same day.

Tuấn grabbed President Phan on the spot.

"Brother Phan! Great to see you here. I have my eye on a 15th-floor river view unit. If you don't look at it now, it'll go to someone else."

Tuấn led President Phan in front of the model of the 15th-floor river view unit. A panoramic view facing the Saigon River tinted by the Ho Chi Minh City sunset. Tuấn was a master at reading the room.

"Brother, look at this view. You don't buy this with money; you grab it with timing."

President Phan's eyes sparkled. Tuấn did not miss that moment. — He intuited that he was hooked.

At the same time, Park Jun-hyuk was quietly conversing with President Phan's wife on the other side.

"Ma'am, by any chance, are you interested in this area because of the children's schooling?"

"Yes, actually, our eldest might be going to RMIT University next year. Thu Duc City is convenient for commuting."

Park Jun-hyuk immediately recorded the new activity in EXAWin:

  • Activity: Model house visit (Accompanied by family)
  • Stage: Solution-Fit (SWV = 2.099)
  • Signal: SA 5.0 (Family accompanied visit = Serious consideration), WA 2.0 (Mentioned RMIT commute convenience)
  • Negation: WN 2.0 (Mentioned price → "It's a bit more expensive than I thought" by wife)

Compound Score Calculation:

S+=max(5.0,2.0)+2.0×0.25=5.5S^+ = \max(5.0, 2.0) + 2.0 \times 0.25 = 5.5 S=2.0S^- = 2.0

Update α, β:

α2=3.0+2.099×5.5=14.545\alpha_2 = 3.0 + 2.099 \times 5.5 = 14.545 β2=1.5+2.099×2.0=5.698\beta_2 = 1.5 + 2.099 \times 2.0 = 5.698 P2(Win)=14.54514.545+5.698=71.9%P_2(Win) = \frac{14.545}{14.545 + 5.698} = 71.9\%

Park Jun-hyuk muttered while looking at the dashboard.

"71.9%... It went up. Being accompanied by family is definitely a strong signal. But the wife's price resistance is concerning. I need to prepare an ROI material for the next meeting."

2.3 Tuấn's Trap: The Anchoring Effect

Tuấn returned from the model house and reported to his team.

"President Phan is hooked. Totally fell for the 15th-floor river view. We can talk down payment next week."

Tuấn's judgment: 95% certain.

However, Tuấn missed three things:

  1. President Phan's wife was feeling burdened by the price.
  2. President Phan was simultaneously looking at a competing project (Vinhomes Grand Park).
  3. The cash flow of his logistics business tightens towards the end of the year.

Tuấn had the opportunity to gather this information. If he had talked to the wife, or asked about the business status. But he was anchored by the single signal of President Phan's "sparkling eyes" and ignored the rest.

It was the trap of intuition.

Chapter 3: 7 Days of Silence

After the model house visit, there was no contact from President Phan for a week.

Tuấn's reaction:

"He must be busy. The logistics business is busy at year-end. Just wait."

Tuấn did nothing. His notebook still read "95% certain." A week passed, and that number didn't change.

Park Jun-hyuk's reaction:

EXAWin's Silence Penalty activated.

silence_gap_days = 5 days. 7 days had elapsed.

n=753+1=1n = \left\lfloor \frac{7 - 5}{3} \right\rfloor + 1 = 1

Unit Penalty = 2.0 × 0.30 = 0.6

Psilence=1×0.6=0.6P_{silence} = 1 \times 0.6 = 0.6 βadj=5.698+0.6=6.298\beta_{adj} = 5.698 + 0.6 = 6.298

Since there was no new meeting yet, the current P(Win):

P(Win)=14.54514.545+6.298=69.8%P(Win) = \frac{14.545}{14.545 + 6.298} = 69.8\%

A yellow warning icon popped up on the dashboard: "Phan Văn Minh — No activity for 5+ days. Follow-up required."

Park Jun-hyuk immediately made a call.

"President Phan, this is Park Jun-hyuk. When you visited the model house last time, your wife mentioned the commute to RMIT. If you need information on school districts in Thu Duc City, should I compile it and send it to you?"

Over the phone, President Phan's voice brightened.

"Ah, Team Leader Park! Actually, I've been a bit busy lately... But that material would be great. And honestly, I've been looking into Vinhomes a bit too..."

Crucial information acquired. Looking into Vinhomes — WN 2.0.

Park Jun-hyuk immediately logged the call details into EXAWin:

  • Activity: Phone follow-up 15 mins
  • Stage: Solution-Fit (SWV = 2.099)
  • Signal: WA 2.0 (Requested school district info = Maintained interest), WN 2.0 (Vinhomes competition)
α3=14.545+2.099×2.0=18.743\alpha_3 = 14.545 + 2.099 \times 2.0 = 18.743 β3=6.298+2.099×2.0=10.496\beta_3 = 6.298 + 2.099 \times 2.0 = 10.496 P3(Win)=18.74318.743+10.496=64.1%P_3(Win) = \frac{18.743}{18.743 + 10.496} = 64.1\%

P(Win) dropped from 71.9% to 64.1%. The negative signal of Vinhomes competition was reflected.

"64.1%... Intuition junkies would still think it's 95%. But the reality is 64%. This is where the real sales begin."

Park Jun-hyuk stayed up all night making a comparative ROI material. Saigon River Sky vs Vinhomes Grand Park. Location, river view premium, expected rental yield, future 5-year value appreciation forecast.

This was the decisive difference from Tuấn. Tuấn thought "I just need to wait," while Park Jun-hyuk received the warning that "the probability is dropping" and moved immediately.


Chapter 4: The Counterattack — The Opportunity Created by Data

4.1 Tuấn's Arrogance

Two weeks later, Tuấn made calls to the 23 A-grade people on his 87-person list. The results were shocking.

  • Out of 23, 7 people had already signed contracts for other projects or lost interest.
  • 5 people didn't answer the phone.
  • Only 11 people were still interested.

Tuấn's A-grade customer retention rate: 48%.

"What is this... They were definitely hooked..."

Tuấn was flustered. But he had no tool to analyze the cause. He just brushed it off, saying "the market has changed."

4.2 Park Jun-hyuk's Precision Strike

At the same time, on Park Jun-hyuk's EXAWin dashboard:

  • Out of 62, P(Win) over 70%: 12 people (Increased from the original 8)
  • P(Win) 50~70%: 18 people
  • P(Win) 30~50%: 22 people (Follow-up required)
  • P(Win) under 30%: 10 people (Effectively abandoned)

As the silence penalty operated automatically, the P(Win) for customers whose interest had cooled naturally declined. Conversely, the P(Win) for customers who provided positive signals through follow-ups increased.

"Lê, create customized investment analysis reports for each of these 12 people. Their points of interest are all different."

  • President Phan: Eldest daughter's RMIT commute + ROI (Residence & Investment combo)
  • President Kim (Korean): Rental yield + Exchange rate risk hedging
  • Doctor Trần: Post-retirement residence + Hospital accessibility

Same condo, different stories. This was the power of data-driven sales.


Chapter 5: Turning Point — President Phan's Decision

5.1 Tuấn's Final Attempt

After 3 weeks of no contact from President Phan, Tuấn finally made a call.

"Brother Phan! It's been a while. I'm still holding that 15th-floor river view unit. But someone else is showing interest..."

Tuấn's strategy was Scarcity Pressure. "If you don't do it quickly, it'll be gone."

President Phan's reply:

"Mr. Tuấn, actually, I've been looking at Vinhomes as well. I'll compare a bit more and contact you."

Tuấn hung up and ground his teeth. Vinhomes. The industry's largest developer. A massive dinosaur that drags customers away on name value alone.

"Can I win against Vinhomes..."

To Tuấn, Vinhomes was a vague terror. He couldn't quantify how dangerous it was, or which way the customer was leaning.

5.2 Park Jun-hyuk's Precision Response

In Park Jun-hyuk's EXAWin, President Phan's current status was displayed precisely:

  • P(Win) = 64.1% → Still leading, but trending downward
  • Silence Penalty Accumulating → β increases every 3 days
  • Competitor Present → Vinhomes recorded as WN 2.0
  • Evidence Maturity: 🌿 Growing → Additional evidence needed

"64%... There's still a chance. But if I stay still, it'll drop below 50%."

Park Jun-hyuk sent President Phan a customized investment comparison report.

📊 Saigon River Sky vs Vinhomes Grand Park Comparative Analysis

ItemSaigon River SkyVinhomes Grand Park
Price per pyeong45 million VND52 million VND
River View Unit Ratio78%32%
Distance to RMIT2.8km (7 mins)8.5km (25 mins)
Expected Rental Yield6.8%/yr5.2%/yr
Move-in DateQ4 2027Q2 2028
Premium Interior OptionIncludedSeparate (Additional 800m VND)

And he added one line at the end:

"President Phan, I respect your choice. My role is to provide you with information so that whatever decision you make is the best one."

Two days later, a call came from President Phan.

"Team Leader Park, I took a good look at that comparison material. My sister-in-law looked at it too... and honestly, Saigon River Sky seems better from an investment perspective. Could we meet again?"

Immediately recorded in EXAWin:

  • Activity: Customer-requested meeting (Customer contacted first)
  • Stage: Proposal (SWV = 2.609)
  • Signal: SA 5.0 (Customer-led re-contact), SA 5.0 (Positive after sharing with family), WA 2.0 (Agreed on investment perspective)

Compound Score:

S+=max(5.0,5.0,2.0)+(5.0+2.0)×0.25=5.0+1.75=6.75S^+ = \max(5.0, 5.0, 2.0) + (5.0 + 2.0) \times 0.25 = 5.0 + 1.75 = 6.75 α4=18.743+2.609×6.75=36.354\alpha_4 = 18.743 + 2.609 \times 6.75 = 36.354 β4=10.496+0=10.496\beta_4 = 10.496 + 0 = 10.496 P4(Win)=36.35436.354+10.496=77.6%P_4(Win) = \frac{36.354}{36.354 + 10.496} = 77.6\%

77.6%. Approaching the threshold for the Proposal stage (T=0.80).

A smile spread across Park Jun-hyuk's face.

"Almost there."


— End of Part 1 —

To be continued in the next part: President Phan's final decision, the fateful showdown between Tuấn and Park Jun-hyuk, and the conclusion of the 480-unit pre-sales war.

Bayesian EXAWin-Rate Forecaster

Precisely predict sales success by real-time Bayesian updates of subtle signals from every negotiation. With EXAWin, sales evolves from intuition into the ultimate data science.

Comments0

Deep Dive

More in Bayesian

Explore Archive
BA04-2. [Novel] Probability in Saigon — The Day Data Beat Intuition (Part 2)
Bayesian
Novel
EXA Intel

BA04-2. [Novel] Probability in Saigon — The Day Data Beat Intuition (Part 2)

The conclusion of the 480-unit condo pre-sales war in Ho Chi Minh City. President Phan's contract, Tuấn's awakening, and the turnaround led by Park Jun-hyuk's EXAWin. The showdown between intuition and data finally reaches its conclusion.

ANALYSIS
BA111. Dynamic Buffers and Backward Scheduling: How to Reconfigure the Factory Around Due Dates
Bayesian
Production Scheduling
EXA Intel

BA111. Dynamic Buffers and Backward Scheduling: How to Reconfigure the Factory Around Due Dates

This story depicts the process of resolving chronic chaos on the manufacturing floor through EXA's advanced Bayesian algorithm and production scheduling engine. Moving away from the indiscriminate push-style production methods of the past, it introduces data-driven simulation and backward scheduling to precisely control process bottlenecks. Through real-time data learning, the system sets dynamic buffers and reorders priorities toward optimizing schedules based on bottleneck process capability for due-date compliance rather than simple utilization. As a result, by suppressing unnecessary WIP and securing protective capacity, the factory undergoes an innovative transformation in which profitability and due-date hit rate rise even while physical machine operating time decreases. It shows the completed form of a demand-driven Pull production system realized by combining human intuition with cold data computation.

ANALYSIS
BA024. The Evolution of EXAWin Bayesian Engine: The Day Data Tuned Its Own Parameters
Bayesian
Auto-Tuner
EXA Intel

BA024. The Evolution of EXAWin Bayesian Engine: The Day Data Tuned Its Own Parameters

The EXA Bayesian Engine calculated win probabilities, but its precision depended on manually configured initial parameters. When 100 historical deals accumulated, the engine was ready to evolve on its own. Grid Search, MCMC Ensemble Sampling, and Cross-Validation — three mathematical pillars working in concert to find optimal parameters. Told as a story.

ANALYSIS
BA04-1. [Novel] Probability in Saigon — The Day Data Beat Intuition (Part 1) | EXA Enterprise