Assumptions and Methodology

Data Cleaning Summary

33 ambiguous names (common names like ASHISH KUMAR, SANDEEP KUMAR, NIDHI, etc. appearing with different services in the same year) were excluded from all cross-year matching. Full list documented in the accompanying Excel workbook.

Chapter OneThe Examination Funnel — How Many Apply, Write, and Get Selected?

1.1 Year-Wise Examination Statistics

Sources: PIB press releases, UPSC Annual Reports.

1.2 Stage-Wise Elimination Probabilities

The Preliminary Examination eliminates approximately 97–98% of candidates who appear. Once a candidate reaches the interview stage, the selection rate improves to roughly one-in-three — the Mains examination is the true differentiator.

1.3 Parsed Allocation Data

The 3,303 usable allocations across the four analysis years represent 2,939 unique individuals. An additional 673 usable individuals in the 2020 reference year provide the baseline against which “Selected at Least Once” status is determined.

Chapter TwoThe “Selected at Least Once” Phenomenon — Repetition Across Years

2.1 Year-Wise Composition: First-Time Selectees vs. Selected at Least Once

For each analysis year, a candidate is classified as “Selected at Least Once” (SAO) if their name appears in any prior year’s allocation list (starting from 2020). Only usable (non-ambiguous) names are classified.

The proportion of SAO candidates has grown steadily from 8.9% in 2021 to 20.8% in 2024. By 2024, approximately one in five usable selectees had already been selected in at least one prior year.

Note: The 2021 SAO figure (8.9%) is based on only one year of lookback (2020). The true SAO proportion in 2021 is likely higher, as candidates selected before 2020 are invisible to our detection. The 2024 figure (20.8%), with four years of lookback, is the most complete and reliable.

2.2 Unique Candidate Summary (2021–2024)

2.3 Frequency of Re-Selection (Within 2021–2024)

Total re-selection instances (usable allocations minus unique individuals) = 364 across the four analysis years. These 364 “seat-opportunities” were consumed by candidates who had already been allocated a service at least once before.

2.4 Year-Pair Overlap Matrix (Usable Names Only)

2.5 Implications

Chapter ThreeAdjacent-Year Re-Selection — The Immediate Return Cycle and Service Improvement Probability

This chapter examines candidates selected in year Y who re-appear in year Y+1 — the most consequential form of re-attempting. All figures are from actual name-matching on usable (non-ambiguous) names only.

3.1 Adjacent-Year Re-Selection Rates

Approximately 8.5–11.3% of each year’s selectees re-appear in the very next cycle, consuming 7.3–10.9% of the next year’s seats. The rate has been trending upward.

3.2 Where Do Adjacent-Year Re-Selectors Come From?

3.3 Service Improvement Probability

~58–68% of adjacent-year re-selectors achieve a service upgrade. But ~21–31% remain in the same tier, and ~7–11% move to a different tier in the opposite direction.

3.4 Service Transitions in the Immediate Next Year — Aggregated Across All Four Adjacent Pairs

The table below shows where candidates moved when re-selected in the very next examination cycle (Y → Y+1 only). This captures only the immediate one-year return pattern.

The most frequent transition pathways: Central Services → IAS (68), Central Services → IPS (61), IPS → IAS (36) — accounting for 165 of 186 total service changes to a higher-ranked group.

3.5 IAS Selection Rate Among Re-Selectors

One-third to two-fifths of adjacent-year re-selectors are allocated IAS in their subsequent selection — a proportion notably higher than the ~18% share IAS represents in the overall allocation.

3.6 The Non-Adjacent Gap: Skipping a Year

Substantial numbers skip a year and return — the re-attempt cycle is a persistent, multi-year phenomenon.

3.7 Summary

Chapter FourService Selection Probability — First-Time Selectees vs. Selected at Least Once

4.1 Service-Wise Breakdown by Year

CSE 2021

CSE 2022

CSE 2023

CSE 2024

4.2 Proportion of SAO Candidates in IAS Over Time

By 2024, nearly four out of every ten IAS seats (among usable names) went to candidates who had been selected in a prior year. The steady acceleration from 13.7% to 37.6% over four years is consistent.

4.3 Lifetime Service Trajectory of All 456 SAO Candidates — First Selection to Most Recent (Full 2020–2024 Window)

Unlike Section 3.4 (which tracked only adjacent-year Y → Y+1 transitions), this section traces each SAO candidate’s entire journey from their first-ever selection to their most recent selection across the full five-year window. A candidate selected in 2020, 2022, and 2024 is represented here as a single trajectory from their 2020 service to their 2024 service — regardless of what happened in between.

Among the 456 SAO candidates (usable, 2+ selections, present in 2021–2024):

163 of 456 SAO candidates (35.7%) ultimately ended up in IAS. This means that nearly two-thirds of SAO candidates did not achieve the IAS allocation despite being selected multiple times.

4.4 What This Means for a First-Time Selectee Targeting IAS

Chapter FiveThe UPSC CSE 2026 Notification — Restricting Re-Attempts by Serving Officers

5.1 What Changed

The UPSC CSE 2026 notification (February 4, 2026) introduced landmark changes:

5.2 Seats Not Available for First-Time Selectees Due to “Selected at Least Once” Candidates

Our cleaned data shows that 20.8% of 2024 selections and 37.6% of IAS seats went to SAO candidates. The adjacent-year data (Chapter 3) shows ~8.5–11.3% of selectees immediately re-appear, with ~66–76% from Central Services and ~19–32% from IPS. Each repeated selection creates a cascading displacement chain through the service ladder.

5.3 Projected Impact

If the 2026 rules had been in effect, approximately ~125 additional seats per year (averaging across the four analysis years) would have been freed for first-time selectees. For IAS, approximately ~44 additional first-time selectee seats per year — a meaningful improvement in selection probability.

5.4 The Debate

Proponents argue for a more level playing field for first-time selectees. Critics counter that restrictions may limit legitimate career choices. UPSC’s stated rationale: administrative efficiency, training resource optimisation, and fairness to new entrants.

Chapter SixThe Economics of Repetition — Opportunity Costs, Near-Term Salary Loss, and Superannuation Penalties

Engaging in repeated examination cycles carries a profound financial penalty — the opportunity cost. This manifests in two devastating phases: the immediate hemorrhage of near-future earnings, and the compounding destruction of terminal wealth and pension benefits at superannuation.

6.1 The Financial Architecture of the Civil Services (7th Pay Commission)

Promotions are bound to years of service, not age or rank. This makes the year of entry the single most consequential financial variable in an officer’s career.

Gross monthly at entry scales to ₹90,000–₹1,00,000 with DA, HRA, and TA.

6.2 The Near-Term Salary Haemorrhage

6.3 The Mathematics of Compounding: Wealth Destruction Through Delay

Candidate A — Enters workforce at 23. Invests ₹40,000/month (10% return) for 5 years (age 23–28), then stops adding money.

Candidate B — Prepares full-time ages 23–28. Zero income. Enters service at 28.

Candidate A’s ₹24 lakh grows to over ₹5 crore purely from time-in-market. The five years of lost compounding destroys roughly ₹2 crore from this tranche alone.

6.4 Full Career Compounding Impact

6.5 The Superannuation Penalty — The Career Ceiling Effect

Because pensions are derived from the last drawn salary, years spent preparing for exams directly amputate the highest-paying years — a lifelong financial penalty extending into retirement. The pension differential between Level 15 and Level 17 is approximately ₹40,000–₹65,000 per month, or ₹5–8 lakh per year. Over a 20–25 year retirement, this exceeds ₹1–2 crore.

6.6 The Private Sector Comparison

A premier-institution graduate taking a private sector job at 23, investing ₹4 lakh/year at 10% for 37 years with 8% annual increment, accumulates approximately ₹12–15 crore — comparable to or exceeding civil services, with more career flexibility and no examination stress. The point is not that private employment is superior — civil services offer irreplaceable public impact and constitutional authority — but the financial cost of delay is substantial, mathematically irreversible, and almost universally underestimated.

6.7 Total Cost of a 5-Year Delay

The years between 22 and 28 are the most financially consequential years of an officer’s career — they sit at the base of the compounding curve.

Chapter SevenWhy Candidates Should Give Their Best Shot at the First Attempt — or the Repeated First Attempt

7.1 The Data Makes the Case

The analysis converges on a single insight: the examination rewards decisive, well-prepared attempts over prolonged, incremental re-tries.

7.2 The Odds Do Not Improve with Repetition

Syllabus evolution: Current affairs are entirely new each cycle. Additional years add the burden of staying current alongside deepening existing knowledge.

Diminishing returns: The first 6–12 months yield the steepest improvement. Subsequent years show diminishing marginal gains.

Psychological erosion: Self-doubt, comparison anxiety, and burnout can reduce performance in subsequent attempts, creating a vicious cycle.

7.3 The “Repeated First Attempt” Philosophy

The most successful candidates treat every attempt as if it’s their only one — 100% preparation intensity each cycle. The data supports this: 88.3% of unique individuals in our analysis years appear only once in the allocation lists.

7.4 The Changing Regulatory Landscape

CSE 2026 introduces strict re-attempt restrictions. The “join a lower service, upgrade later” strategy now carries: mandatory resignation from CSE 2028 onward, loss of seniority, a single one-time improvement window, and the psychological burden of having something to lose.

7.5 The Adjacent-Year Evidence (Chapter 3 Revisited)

Even among the most determined re-selectors — those returning in the very next year — only ~58–68% achieve a service upgrade. ~33–42% are allocated IAS in the subsequent selection. The remaining ~32–42% stay in the same tier or move to a different one.

7.6 A Framework for Decision-Making

Before re-attempting, evaluate honestly:

Marginal benefit: Is the target service significantly different in career satisfaction and lifestyle? IPS, IFS, and Central Services offer deeply fulfilling careers.

Marginal cost: A 3-year delay costs several crores in terminal wealth. A 5-year delay can destroy ₹4–17 crore in lifetime financial value — including lost compounding, reduced terminal pay, and diminished pension.

Marginal probability: Of 456 SAO candidates, only 163 (35.7%) ultimately landed in IAS — nearly two-thirds did not achieve the IAS allocation despite multiple selections.

Changing rules: CSE 2026 norms introduce resignation risk, training disruption, and irreversible administrative consequences.

7.7 The Bottom Line